The Settlement Issue in Turkey and the Kurds
Social, Economic and Political Studies of the Middle East and Asia (S.E...
112 downloads
2424 Views
4MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
The Settlement Issue in Turkey and the Kurds
Social, Economic and Political Studies of the Middle East and Asia (S.E.P.S.M.E.A.) (Founding editor: C.A.O. van Nieuwenhuijze)
Editor
Dale F. Eickelman Advisory Board
Roger Owen (Harvard University) Yann Richard (Sorbonne Nouvelle)
VOLUME 102
The Settlement Issue in Turkey and the Kurds An Analysis of Spatial Policies, Modernity and War
by
Joost Jongerden
LEIDEN • BOSTON 2007
This publication has been made posssible by the financial support from the Netherlands Organization for Scientific Research (NWO). On the cover ( from top to bottom): 1) view on an old part of the city, Melik Ahmet Quarter, Diyarbakır; 2) view on an old part of the city, Lalebey Quarter, Diyarbakır; 3) returnees to an evacuated and destroyed village in the province of Diyarbakır; 4) view on Bağlar, Diyarbakır. 5) election meeting of DEHAP in Diyarbakır. (Photographs by the author, nos. 1–4: 2005; no. 5: 2002.) English editing: Andy Hilton This book is printed on acid-free paper.
ISSN 1385-3376 ISBN 978 90 04 15557 2 © Copyright 2007 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. printed in the netherlands
CONTENTS List of Tables .............................................................................. List of Graphs ............................................................................ List of Boxes ............................................................................... List of Terms and Abbreviations ............................................... Notes on Turkish and Kurdish Spelling and Pronunciation ..... Note on Style .............................................................................. Foreword ..................................................................................... Preface ......................................................................................... Chapter One
vii ix xi xiii xvii xix xxi xxv
Introduction ......................................................
1
Chapter Two Soldiers and Settlements Turkish Counter-Insurgency and the Strategy of Environment Deprivation ....................................................
43
Chapter Three Settling the Southeast The Rise and Fall of a Rehabilitation Doctrine ...........
93
Chapter Four The Integrated Settlement Network Towards a New Spatial Organization of the Countryside ....................................................................
119
Chapter Five Settlement Issue and Settlement Activities The Colonization of Peoples and Territories .................
173
Chapter Six Claiming the Land Empirical Observation on the Ground .............................
219
Chapter Seven
Summary and Conclusions ............................
281
Annex: Maps of the Pilot Projects .............................................
321
Bibliography ................................................................................
335
General Index .............................................................................
349
LIST OF TABLES Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 2.8 Table 3.1 Table Table Table Table Table Table Table Table
3.2 4.1 4.2 4.3 4.4 4.5 4.6 4.7
Table 4.8 Table 4.9
Table 4.10
Distribution of village guards in 2003 ................. Structure of Turkish army principal tactical units before and after 1992 reorganization .......... Number of evacuated and burned villages 1991–2001 ............................................................. Population growth in OHAL and Kurdistan provinces in Turkey ............................................... Proportion of the population living in rural areas ...................................................................... Cumulative population growth in Diyarbakır districts .................................................................. City and village population and annual growth rate of population in the Kurdistan region by province ................................................................. City and village population and annual growth rate of population in the Southeast by province ................................................................. Districts and number of sub-districts, villages and hamlets in Van ............................................... Center-village pilot projects .................................. The ranking of settlements in Turkey ................. Population, villages, and hamlets in Turkey ......... Number of villages and hamlets per province ..... List of Governments ............................................. List of Presidents .................................................. Regions in Turkey ................................................. Number of districts in the provinces of the Kurdistan area in Turkey ...................................... Number of villages, hamlets and center-villages per province .......................................................... Number of center-villages by region and province according to the decision of the Council of Ministers in 1983, and number of villages and hamlets according to a 1990 count ...................... Provinces with the highest absolute number of center-villages according to a 1990 count ............
66 69 82 86 86 87 88 90 103 113 123 129 132 135 138 140 142 159
162 163
viii Table 4.11 Table 4.12 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7
list of tables Provinces with above average settlements per center-village ......................................................... Provinces with the highest absolute number of hamlets .................................................................. Figures related to village return in the province of Diyarbakır as of January 1, 2004 ................... Characteristics of selected villages ....................... Composition of the populations of the city of Diyarbakır in 1518 and 1540 ............................... Population of the city of Diyarbakır ................... Composition of the population of Diyarbakır according to the Salname of 1896–1905 ............. Population of İslamköy before evacuation in 1994 ....................................................................... Resettlement of Turkish Bulgarians in Diyarbakır .............................................................
164 165 225 229 233 234 236 257 271
LIST OF GRAPHS Graph Graph Graph Graph Graph
2.1 4.1 4.2 4.3 5.1
Internal displacement in Turkey ........................... Population density .................................................. Regions in Turkey .................................................. Graphical representation of a village-town .......... Representation of the conventional idea of modernization as a process of transition (from ‘traditional’ to ‘modern’ society) .................. Graph 5.2 Representation of modernization as a process of differentiation (between ‘traditional’ and ‘modern’ society) .................................................... Graph 5.3 Representation of an alternative modernization, as a synthesizing process in the rural/urban and agricultural/industrial matrix ................................ Graph 6.1 Genealogical tree of the Mala Süleyman .............
81 131 139 169 213 214 214 274
LIST OF BOXES Box 1.1 The Borders of Turkey as Determined by the Treaty of Peace with Turkey, Signed at Lausanne, July 24, 1923 .............................................................. Box 2.1 Resettlement in Algeria .............................................. Box 2.2 Schemes ...................................................................... Box 3.1 Destruction, Reconstruction and Legislation ............ Box 4.1 Extract from Newspaper Article by Mustafa Ok, 1962 (on the abolition of all small hamlets and villages and the concentration of their population in large rural settlements) ........................................... Box 5.1 Specification of Areas in the Southeast where it was Deemed Desirable to Increase the Density of the Culturally Turkish Population (Zone 1) .............. Box 5.2 Populations Movements to and from Anatolia between 1912 and 1924 ............................................. Box 5.3 Nusret Kemal Köymen .............................................. Box 5.4 Urban and Rural Modernization in the 1930s: The case of Kadro and Ülkü .................................... Box 6.1 İbrahim Pa a and Diyarbakır .................................... Box 6.2 The Village Guards of İslamköy ...............................
12 77 82 115
153 176 182 196 202 245 261
LIST OF TERMS AND ABBREVIATIONS Ala Rizgarî ADYÖD
Flag of Liberation (political party) Ankara Democratic Higher Education Association (Ankara Demokratik Yüksek Öğrenim Derneği) AKP Justice and Development Party (Adalet ve Kalkınma Partisi) ANAP Motherland Party (Anavatan Partisi) ARGK Kurdistan People’s Liberation Army (Artê a Rizgariya Gelê Kurdistan) belediye municipality cazibe merkezi attraction center (a rural planning concept) CHP Republican People’s Party (Cumhuriyet Halk Partisi) çiftlik rural settlement type, composed of a few (arable) farms (smaller than a mezra) DAP Eastern Anatolia Project (Doğu Anadolu Projesi) DDKO Revolutionary Eastern Cultural Hearths (Devrimci Doğu Kültür Ocakları) Dev-Genç Revolutionary Youth DİE State Statistics Institute (Devlet İstatistik Enstitüsü) Doğu ve Güneydoğu Anadolu Bölgesi Köye Dönü ve Rehabilitasyon Projesi Alt Bölge Geli me Planı East and Southeast Anatolia Return to Village and Rehabilitation Project Sub-Regional Development Plan dönüm unit of square measure used for land (equivalent to about 1,000m²) DPT State Planning Organization (Devlet Planlama Te kilatı) DSP Democratic Left Party (Demokratik Sol Partisi) DYP True Path Party (Doğru Yol Partisi) GAP Southeast Anatolia Project (Güneydoğu Anadolu Projesi) Göç-Der Migrants Social Relief and Culture Association (Göç Edenler Sosyal Yardımla ma ve Kültür Derneği) halk evleri people’s houses people of the same place of origin (village, district or hem ehri region) HRK Kurdistan Liberation Forces (Hêzên Rizgariya Kurdistan) IDP Internally Displaced Person IHD Human Rights Association (İnsan Hakları Derneği ) kaymakam head official of a district KHGM General Directorate of Rural Services (Köy Hizmetleri Genel Müdürlüğü) KHRP Kurdish Human Rights Project Kürdistan Devrimcileri Kurdistan Revolutionaries Hamidiye Alayları the Hamidiye Regiments, irregular tribal forces established by Sultan Abdulhamid II, which existed 1891–1909
xiv HRW il ilçe İskan Kanunu Kadro KDP KİP kom köy
köycülük köycülük cemiyeti Köy Kanunu köy-kent Köy Korucuları KUK mahalle Mazlum Der MEGEV merkez-köy mezra MİT MGK
muhacir muhtar nahiye oba
list of terms and abbreviations Human Rights Watch province district Settlement Act (Law Number 2510, passed in 1934) Cadre, a journal published 1931–1935 Kurdistan Democratic Party (Partiya Demokrat Kurdistan) Kurdistan Workers Party (Kürdistan I çi Partisi); note: to be distinguished from PKK, which means the same. permanent settlement type to be found in mountainous areas, based on animal husbandry i) a settlement with less than 2,000 inhabitants living compactly or dispersed, who jointly own a mosque, school, pasture, summer pasture and part of a forest reserved for fire-wood from a village, together with their vineyards, gardens, and fields (Turkish legal definition) ii) a rural settlement with an economic structure based mainly on agricultural activities, face-to-face relations between its members, and little specialization of labor (Turkish sociological definition) peasantism peasantist community Village Act (passed in 1924) village-town (a rural planning concept) Village Guards, an anti-PKK paramilitary force established in 1987 Kurdistan National Liberators (Kürdistan Ulusal Kurtulusçuları) i) neighborhood, quarter (of a village/town/city) ii) small rural settlement, between a hamlet and village in size Association for Human Rights and Solidarity with the Oppressed (İnsan Hakları ve Mazlumlar için Dayanı ma Derneği) Mesudiye Development Foundation (Mesudiye Geli me Vakfı) center-village (a rural planning concept) i) hamlet, ii) small rural settlement type based on arable farming National Intelligence Agency, the Turkish secret service (Milli İstihbarat Te kilatı) National Security Council, an executive body composed of top military commanders and government ministers that ‘advises’ the government on matters of state (Milli Güvenlik Konseyi) Muslim immigrant from the Balkans, Caucasus, or Crimea village chief, a standard position in Turkish society administrative division between village and town, composed of a cluster of small rural settlements and villages (Ottoman term) semi-permanent settlement type to be found in mountainous areas, based on animal husbandry
list of terms and abbreviations OHAL Özel Tim
xv
State of Emergency (Olağanüstü Hal) Special Team (under authority of the Gendarmerie) Özel Hareket Timi Special Action Team (under authority of the police) PFLP Popular Front for the Liberation of Palestine PFLP-GC Popular Front for the Liberation of Palestine— General Command PKK Kurdistan Workers Party (Partiya Karkerên Kurdistan) PPKK Vanguard Party of Kurdistan Workers (Partiya Pê eng a Karkerên a Kurdistan) PPSF Palestinian Popular Struggle Front PSK Socialist Party of Kurdistan (Partiya Socialîsta Kurdistan) Rizgarî Liberation (a political party) RP Welfare Party (Refah Partisi) rurban a hybrid settlement type merging village and city sancak district (Ottoman term) tarım kent agricultural town (a rural planning concept) Teko în Struggle (a Kurdish political party) TESEV Turkish Economic and Social Studies Foundation (Türkiye Ekonomik ve Sosyal Etüdler Vakfı) TİHV Human Rights Foundation of Turkey (Türkiye İnsan Hakları Vakfı) tımar fief, person held for military service (Ottoman term) TİP Workers Party of Turkey (Türkiye İ çi Partisi) TODAİE Public Administration Institute for Turkey and the Middle East (Türkiye ve Orta Doğu Amme İdare Enstitüsü) toplu çiftlik collective farm (a resettlement project) toplu kondu collective shelter (a resettlement project) TSBD Turkish Social Sciences Association (Türk Sosyal Bilimler Derneği) TSK Turkish Armed Forces (Türk Silahlı Kuvvetleri ) TKV Turkish Development Foundation (Türkiye Kalkınma Vakfı) TTK Turkish History Foundation (Türk Tarih Kurumu) Türk Ocakları Turkish Hearths (a youth organization) UNHCR United Nations High Commissioner for Refugees Ülkü (Ülkü Halkevleri Mecmuası) Ideal, a journal published in the 1930s vali governor of a province yayla summer pasture
NOTES ON TURKISH AND KURDISH SPELLING AND PRONUNCIATION The names of Turkish/Kurdish people and places are given in the Turkish/Kurdish forms in this book, employing the standard Turkish/ Kurdish spelling systems. The Latin letters in the Turkish system are pronounced more or less the same as their English equivalents, with the following exceptions: Letter English pronunciation a c ç ğ ı j ö u ü v
generally softer than the English, like the a in bah (rather than hat) j, as in jam ch, as in church lengthens preceding vowel; thus ağa is pronounced a-a like the a in gentleman like the s in measure like the ir in bird sh, as in ship like the u in put like the o in do softer than the English
The Latin letters in the Kurdish system are pronounced more or less the same as their English equivalents, with the following exceptions: Letter English pronunciation a c ç e ê i
a, as in father j, as in jam ch, as in church phonetic realization varies from the e in bed to the a of French passé like the é in French passé neutral, unemphasized vowel, as the e in paper
xviii
notes on turkish and kurdish spelling and pronunciation
î j
like ea in please is pronounced as in French sh, as in ship like the u in church like the oo in food like the ch in Dutch or German acht like the w in war
u û x w
NOTE ON STYLE Italics are used to show foreign language (i.e. non-English) terms, emphasis, and reference to terms (e.g. . . . the Armenian word for monastery . . .). Inverted commas are reserved for quotations and to indicate doubt, novel terms, etc. (e.g. . . . the so-called ‘return’ of inhabitants . . .). Although it is more conventional to show references to terms with inverted commas than italics, the difference between foreign language terms, emphasis and reference is probably clearer than that between reference and doubt, novel terms, etc., at least in this work, which is the reason why the convention is altered here.
FOREWORD Martin van Bruinessen Utrecht University and ISIM* For fifteen years, from 1984 to 1999, the eastern provinces of Turkey (or, if one wishes, northwestern Kurdistan) were ravaged by a guerrilla war that pitted the Kurdish insurgent movement PKK against the full force of the Turkish state. Around 35,000 people lost their lives in the conflict, most of them Kurds. Hundreds of thousands of Kurdish villagers were forced to leave their homes and land and resettle elsewhere in Turkey. In the 1990s, over 3,000 settlements were forcibly evacuated and destroyed by the army, as part of an effort to deny the guerrilla bands in the mountains access to food and civilian support and to isolate them completely from the civilian population. In many of these cases, the villagers had been given the choice between signing up as ‘village guards’ (korucu) and giving up the village. The ‘village guards’ are irregular militia units, commanded by a local chieftain but supervised by the nearest army post, who are paid and armed by the government to keep the PKK away from their district. They have also been obliged to take part in large-scale military operations against guerrilla units further away from their village. In several regions, the only inhabited villages left were korucu villages. As Kurdish allies of the Turkish armed forces, the korucu enjoyed virtual immunity and could use their arms for the exercise of private violence as well. They revived old feuds and took revenge at old enemies, killed and looted, and took by force the land, property and women that they desired. The korucu system resulted in a sharp increase in intraKurdish conflicts, feuds between tribes or villages, causing many more families to leave the region. The number of korucu on the government payroll increased steadily through the 1990s, reaching 65,000 by the end of the decade. After the capture and trial of the PKK chairman, Abdullah Öcalan, in 1999 and in response to Öcalan’s publicly renouncing the armed
* International Institute for the Study of Islam in the Modern World.
xxii
foreword
struggle, the party leaders who were still at large announced an end to the hostilities and ordered their fighters to avoid further clashes. The guerrilla groups that still were active in Turkey disbanded, surrendered or withdrew to Iraqi Kurdistan. After some time, the military downgraded its state of alert, and the government lifted the state of emergency, which was one of the preconditions for Turkey’s accession to the European Union. The region was gradually returning to normal life, and many of the villagers who had involuntarily left started exploring the possibilities of their return to their villages. There were families who returned to their destroyed village and took up cultivating the land, without daring yet to rebuild their homes. Here and there, people provisionally restored houses and prepared to resettle in the village. In many localities, the military authorities prevented resettlement or even return. Elsewhere, former inhabitants found that their land had been appropriated by korucu, who resisted their return. Economically, the revival of agrarian life in the Kurdish region was highly desirable, but the government preferred the resettlement of the returnees in new, larger and more concentrated settlements, where better services could be provided and closer surveillance was possible. It was this process of return to the villages and rehabilitation of the war-torn region that Joost Jongerden set out to study in 2001. Joost had been a country expert for Amnesty International during the preceding years and was well-informed on the political situation in Turkey and the shifting policies of both the PKK and the government. The forced village evacuations were an issue he was familiar with due to his work at Amnesty. He was moreover trained as a rural sociologist in a department (at Wageningen University) where post-conflict reconstruction was a chief focus of attention. Thus he was well equipped for this research project; and he embarked upon it at the right time to actually observe the onset of return and reconstruction. He traveled extensively in the region, visiting both villages where spontaneous remigration had started and government-run ‘centre-village’ resettlement projects. Access to several of the evacuated regions remained restricted, and like many would-be returnees, Joost was repeatedly turned back at military checkpoints. But he succeeded in visiting many villages where resettlement was taking place and he provides us in this book with a detailed first-hand account of social dynamics and conflicts in the process. He is, to my knowledge, the only scholarly observer of this process of return to the village and the efforts to rehabilitate the Kurdish rural districts. That alone makes this a unique and significant
foreword
xxiii
study. He adds moreover important new perspectives by placing his study in the broader historical context of population displacements in the region and the Turkish government’s resettlement policies since the establishment of the Republic. The village evacuations of the 1990s were not the first massive displacement of populations in the region, nor even the largest. The obvious earlier example is the massive deportation of all Armenians, followed by systematic slaughter, during the First World War. Many Kurds perceived a parallel between what had been done to the Armenians and the evacuation of Kurdish villages of the 1990s, although in the latter case there were no large-scale massacres. The Ottoman Empire had moved and resettled population groups over great distances as a means of territorial control, and the Republic of Turkey has also resorted to that policy. Kurdish uprisings in the 1920s and 1930s were punished by massive deportations to western Turkey, and Muslim ethnic groups originally from the Caucasus or the Balkans were resettled in the Kurdish region. Many settlements show the traces of these and earlier population movements, in their physical structure and material culture, in folklore, in their names, and in the memories of their inhabitants. The four villages studies in this book illustrate that very well. Many towns and villages in the region have been officially renamed under the Republic, some even twice, in order to erase their non-Turkish character, but people continued using the old names. Some of the new Turkish village names, like İslamköy, moreover appear to give away the fact that the previous inhabitants were non-Muslim. The Armenians are very much present today in people’s memories, in fact even more so than I found to be the case thirty years ago. Jongerden’s case studies bring out clearly the complexity of the settlement issue and of the ethno-history of the region and thereby provide an important nuancing of earlier narratives of wholesale deportation and return. Jongerden’s micro-level studies of resettlement are complemented by a fascinating history of the attempts by planners and governments to come to grips with Turkey’s extremely dispersed rural settlement structure. This has probably been the country’s major and most untractable problem of governmentality. It has been observed of urban squatter settlements ( gecekondu), which have expanded and consolidated themselves in contravention of all efforts at urban planning, that they give the lie to the claim that the Turkish state is strong, and society powerless vis-à-vis that state. In spite of heavy-handed measures of forced
xxiv
foreword
migration, the rural settlement pattern shows even more clearly the limits of the power of the state. Nation-building and modernization demanded that the rural population be educated and otherwise disciplined, but the sheer number of settlements, their low average size, and difficulties of terrain made this an impossible task. From the early republican period on, planners of different ideological orientation have considered a drastic transformation of the rural settlement pattern, concentrating the population in much larger and more easily accessible units, as an essential measure. But the costs of such an operation are so forbidding that the plans were never implemented, apart from a few very recent pilot projects, as detailed in this book. In tracing the genealogy of the ‘center-village’ concept to its different origins in rural development theories as well as counter-insurgency strategies, Jongerden makes a fascinating contribution to the history of state-society relations in Turkey and adds an important perspective to the history of the Kurdish conflict. The thinking of those who ‘see like a state’ in Turkey is meaningfully juxtaposed to the view from below in the village studies. Jongerden offers important new insights in both. This book will find recognition as one of the important studies of Turkey and Kurdish society at the turn of the millennium.
PREFACE Heaven cried out, earth groaned Day grew silent, darkness emerged Lightning flashed, fire broke out [Flames] crackled, death rained down Gilgamesh, Tablet IV We must cease once and for all to describe the effects of power in its negative terms: it ‘excludes’, it ‘represses’, it ‘censors’, it ‘abstracts’, it ‘masks’, it ‘conceals’. In fact, power produces; it produces reality; it produces domains of objects and rituals of truth. Michel Foucault War has become a regime of biopower, that is, a form of rule aimed not only at controlling the population but producing and reproducing all aspects of social life. Michael Hardt and Antonio Negri1
The research for this book was conducted during the period 2002– 2005, although its origins date back several years before that. In the summer of 1996, over the weekend of June 3–4, the United Nations Conference on Human Settlements (Habitat) assembled in İstanbul to ‘endorse the universal goals of ensuring adequate shelter for all and making human settlement safer, healthier, and more livable, equitable, sustainable and productive’. Delegations to the UN Habitat Conference in İstanbul presented their governments’ ‘national reports’ on how to provide adequate shelter, stimulate social development, eradicate poverty, protect the environment and develop the economy, all while adhering to good governance. The National Report and Action Plan prepared by the host country, Turkey, unfolded a plan for the development of a peaceful and stable city life, the improvement and renewal of gecekondu (shanty town) housing, and the provision of adequate shelter for those citizens who had lost their homes in the southeast of Turkey as a result of the twelve-year war between the state and the PKK. The report was purportedly a product of co-creation of nearly 244 organizations, three-quarters of which were non-governmental bodies or civil society organizations. 1
Citations: (Dalley 2000: 69), (Foucault 1977: 194) and (Hardt and Negri 2004: 13).
xxvi
preface
Several civil society organizations, among them the human rights organization İHD (İnsan Hakları Derneği), criticized the National Report and Action Plan because, they argued, it was marked by ‘omissions’. Firstly, the production of this national report on adequate housing and sustainable settlement was characterized by an institutional omission. Although the report was supposedly a product of co-creation between State and civil society organizations, non-governmental organizations claimed that the Turkish authorities had been selective in the process of (genuine) consultation, generally ignoring those organizations which took a critical stand against State policies, and particularly those concerned with the so-called Kurdish issue. The second omission was related to the content of the report, which largely ignored the issue of the destruction of shelter and the mass-displacement of the rural population in the Kurdistan region in southeast Turkey by the Turkish army. Civil society organizations also criticized the UN for organizing a conference on adequate housing and social development in a country that had been systematically destroying settlements and people’s livelihoods. In the shadow of the official UN conference, several civil society organizations organized an alternative habitat conference, in which such issues as the large scale displacement of the rural population in the Kurdistan region of Turkey was to be discussed. As background information, the Human Rights Association İHD distributed a report named ‘The Burned and Evacuated Settlements’ which documented the destruction 2,047 rural settlements in the Kurdistan region of Turkey up until 1996. A debate about such issues was not welcomed by the Turkish authorities, and the opening session of the alternative forum was disrupted by the police, who ordered the delegates to cease their activities and disperse. Although this intervention was severely criticized by delegates in the official conference, sessions of the alternative forum had to be cancelled. Eventually the frightening and violent evacuation of rural settlements could only be discussed in a remote location in the Beyoğlu quarter in İstanbul. I was in İstanbul during this time, participating in some of the sessions of the alternative forum. Together with two colleagues, for three weeks I carried out interviews for a collaborative publication on the displacement of the rural population in the Kurdistan region of Turkey. We interviewed Kurds in İstanbul whose villages had been destroyed and evacuated and who, frankly, had gone through a living hell. These victims of the violent campaign of displacement had settled in new districts on the city outskirts, such as Umraniye, where they crowded
preface
xxvii
in with relatives and constructed shanty dwellings, or they had rented rooms in one of the dilapidated backstreets of an old quarter like Beyoğlu. Among the people we interviewed was an elderly couple in their seventies, occupying one such room in a ramshackle building in Beyoğlu. The man explained how, the year before, in 1995, the Turkish army had forcibly evacuated and then torched his village near the district town of Ovacık in the province of Tunceli (Dersim). He explained to us that this was not the first time that the village in which he lived had been evacuated and destroyed by the military, displacing him and his family. In the summer of 1938, in his early childhood, he had been among the people from Ovacık who were herded together by soldiers to be deported, first to Hozat and later to Elaziğ. From there, the families had been scattered over Turkey, some to Konya, others to Balıkesir, Bolu, or Çankırı. Only after an exile (sürgün) of more than 10 years were he and his family finally permitted to return. ‘How many years’, the man asked, ‘Will it take until we are allowed to return this time?’ Ironically, the issue of people’s return to their villages had entered the political agenda in Turkey already in 1994, one year before the evacuation of the village of the old man and his wife. It reminded me of a poem by Bertolt Brecht in which he wrote that when the rulers talk of peace the common people must be prepared for war. Were the return initiatives of the Turkish government newspeak for evacuation, to give an impression of government action in order to silence critics? Did they fail to materialize because of institutional disarray or disagreement? Or was there another explanation? Whatever the reason, between 1995 and 2000 various return initiatives were announced, including the ‘Village Return Southeast Anatolia Restoration Project’ (Köye Dönü Güneydoğu Anadolu Onarim Projesi) and the ‘East and Southeast Anatolia Village Return and Rehabilitation Project Sub-Regional Development Plan’ (Doğu ve Güneydoğu Anadolu Bölgesi Köye Dönü ve Rehabilitasyon Projesi Alt Bölge Geli me Planı), and yet none came to fruition. Instead, village evacuations continued and the implementation of village return schemes failed to materialize. Relatively few people returned to live again in their villages. Five years after the UN Habititat conference, my visit to İstanbul and the interviews with displaced villagers, the Netherlands Organization for Scientific Research (NWO) awarded me a grant for the study the issue of ‘return’, thereby affording me the opportunity to investigate the questions that had been playing on my mind. I settled in Diyarbakir,
xxviii
preface
the main city in the southeastern region, where a large proportion of the displaced population had settled, on either a temporary (transitional) or permanent basis. In the city, displaced villagers had created their own ‘spaces’ (coffee-houses, grocery shops, etc.), spaces where one could meet the displaced villagers and which made it possible to travel from ‘village’ to ‘village’ in the city. Also I traveled in the area, and did part of my fieldwork in evacuated settlements that had been re-inhabited. In the course of this study I became convinced that the issue of return was part of a larger phenomenon, the production of space, and that war, rehabilitation and (re)development, the constituting elements of return, had to be analyzed from this perspective. This had the effect of expanding the limited study of return into a larger project investigating the production of space by settlement and resettlement projects, of which the issue of return was the immediate concern, but not the only object of study. During this research I found it striking that after decades of village evacuations and displacement and the more than ten years that the issue of return had been on the Turkish political agenda, several studies had been published about return from a legal and human rights perspective, but none from a sociological one. The issue of return as a whole had not been the object of a sociological study, let alone one based on fieldwork. In spite of difficulties and obstacles, I persisted in completing the study and writing this work, partly inspired by a poem by the Kurdish writer, Bêrî Bihar (2002): Kes tûne bû Binivîse pirtûka me. Em bi tevahiya jiyana xwe, Li benda destekî bûn, Ku porên me miz de.
There was nobody To write our book. All our lives We had waited for the hand To caress our head.
Di bin lihêfekê de Deh pê û penc serî. Ti tê me germ dikir, Xu k û biratiya me.
Under one blanket Ten feet, five heads. Warming ourselves With brotherhood.
Li benda çîrokekê man, Di heman jiyana xwe de, Kesî ne got ji me re, Em nebirin xewnên rengîn.
Waiting for a story The whole of our lives, But nobody told us, Nobody took us to colorful dreams.
preface
xxix
This is not a book that will bring anyone colorful dreams. But it does tell an untold story, or at least, takes up an original subject. Many of the displaced and returnees with whom I spoke in my research encouraged me to write their stories. This book could not have been written without these people, people who shared with me some of the darkest periods of their lives. I do not imagine myself to have written the book on the sociology of return, but I do hope to have made a contribution, to have revealed some of the logic of evacuation, displacement, and return, and to have articulated a critique that implies alternatives, alternatives both to the kinds of state policies described and to the policies of the state per se. Finally, instead of creating nightmares, I hope to have written a book that may just help to make some dreams true.
CHAPTER ONE
INTRODUCTION “It is a stupid name enough!” Humpty Dumpty interrupted impatiently. “What does it mean?” “Must it mean something?” Alice asked doubtfully. “Of course it must”, Humpty Dumpty said with a short laugh: “my name means the shape I am—and a good handsome shape it is too. With a name like yours, you might be any shape, almost.” Lewis Carroll1
The Logic of Resettlement More than five decades ago the political scientist Robert Koehl (1953) highlighted the essential logic of resettlement—as the attempt to dominate a region by removing from it all those who are believed to be uncontrollable, and filling it instead with a controllable population. Practices of resettlement, he argued, are intended to master a territory. Koehl evaluated resettlement as a means of cleansing and exchanging populations, and related these resettlement practices to the process of transformation by which empires became nation-states and the consequent concern of these states with the characteristics of their subjects. He argued that as early as the 18th century political theorists had taught that a disciplined, productive population was the true wealth of a sovereign. The goal was to maximize the population, by marriage or conquest, without much regard to people’s (cultural) characteristics. However, at the turn of the 20th century something dramatic occurred. Important European imperial state-actors—the Ottoman, Austro-Hungarian and Russian empires—disintegrated and new states emerged with their population politics modeled on the idea of nationalism, a political concept holding that the borders of political units (state) and cultural units (nation) should coincide and teaching that the power of a state depends on the degree to which its subjects
1
Lewis Carroll, ‘Through the Looking Glass’, 1872.
2
chapter one
respond to the ideal of the particular cultural identity that is thought to characterize the nation (Koehl 1953: 231). Since the borders of states rarely corresponded with those of ‘cultural units’, the conviction that these borders ought to coincide gave rise to the idea of resettlement, an exchange of populations between political-cultural entities employed as a method of bringing logic to the map. The idea of bringing logic to the map comprised the territorial reshuffling of people between states and aimed at overcoming the discrepancy between the nationalist ideal of congruence between political and cultural units and the reality on the ground. This spatial binding of polity and culture—all as if they are instances of the same substance, the nation—marks the modern project of nationalism. Its aim is the attachment of a culture-based political regime to space and, as such, create the territorial belonging of ‘the nation’. Initially, the term resettlement was defined as the physical transition of an individual or group from one society to another, or from one political area to another. It referred to such practices as the so-called transfer of populations between political entities by international agreement (Koehl 1953: 232; Stola 1992: 326), such as the population exchanges agreed between Turkey and Greece in 1923, an inter-state reshuffling which affected more than one and a half million people (and which will be discussed in more detail in chapter 5). The idea of resettlement as a tool for the management of populations aiming at the production of a congruence between state and culture is a modern one and a key proposition of social and political science. It is hold for a functional logic of modernity. In an article on Turkey, the modernist anthropologist, philosopher and sociologist Ernest Gellner argues that resettlement, interpreted as eviction from a particular political-cultural entity, is not only one of the options a state has for the production of congruence, but also an inevitable one. People, he argues, have to be competent in the idiom employed by the educational, economic, and administrative bureaucracies that surround them, and their personal characteristics must be compatible with the self-image of the culture in question. If people master the dominant idiom—and it is an abstract idiom that can be mastered only through formal education—and are acceptable to it on the basis on their personal characteristics, then they can acquire membership of the nation. If not, their life is a series of humiliations. Gellner argues that this basic situation of modernity forces people to be nationalists. If they do not, people have a number of options, from gradual assimilation to
introduction
3
migration and commitment to an alternative, competing nationalism. Society, he says, has similar options regarding those who do not fit its paradigm, from assimilation (assimilation being a two-way process, supposing assimilation of as well as assimilation into), to expulsion and murder (‘ethnic cleansing’). These, Gellner says, are the processes we have been witnessing in the twentieth century and the basic underlying pattern of nationalism, requiring, as it does, a sufficient congruence between state and culture for a viable polity. Resettlement, understood as the deportation of people from one political entity to another, is simply considered a necessary and inevitable condition of modernity. For Gellner, and many others, it is just a fact of modern life (Gellner 1997: 239–240). However, the management of populations and the territorial production of a congruence between political and cultural units, is not only to be described in negatively, in terms of exclusion and deportation, for power is productive: power produces reality, creating domains of objects and rituals of truth (Foucault 1977: 194). A fundamental problem and primary activity of nation-states is the production of the social category of ‘the people’, the production of population. ‘A social formation only reproduces itself as a nation,’ the philosopher Balibar argues, ‘to the extent that through a network of apparatuses and daily practices, the individual is instituted as homo nationalis from cradle to grave, at the same time as he or she is instituted as homo oeconomicus, politicus, and religiosus’ (Balibar 2002: 93). Fostering a population becomes the purpose of administration. The individual is a fabricated reality (Foucault 1977: 194).2 The process of fabricating people as ‘the people’ is analyzed in Eugene Weber’s ‘Peasants into Frenchmen: The Modernization of Rural France, 1870–1914’. Weber not only argues that peasants became Frenchman only in the closing decade of the 19th century, but also describes the means for the production of national identity (Weber 1976: 292–302; 303–18)—for example through the use of ‘education’ and ‘military service’. In this study, we are concerned with an understanding of spatial practices, in particular the practices of settlement and resettlement, as a means for the constitution of social life, focusing on the production of a Turkish national identity and the advancement of the Turkish state-administration.
2 Gellner too seems to acknowledge this, but he takes it as a fact of life, not as one of its basic problems.
4
chapter one The Study of Resettlement
Broadly speaking, one can say that the meaning of the word resettlement is derived from the verb to settle.3 Settlement, the noun form of the verb, is defined as the movement and establishment of people upon the land in some relatively permanent matter. By common usage, the prefix rebefore settle(ment) tends to carry the additional idea of force, implying an enforced movement, a particular kind of dislodgement. Re-settlement thus incorporates a form of de-settlement, with settlement being the more inclusive concept, and resettlement used only where there is a particular form of dislodgement before settlement (Chambers, 1969: 11). Whereas resettlement is essentially about leaving and ejection, settlement focuses on arriving and integration. The difference is one of perspective, and, because this is about tendency and implication, there is not a clear demarcation between the two. Both settlement and resettlement involve a planned, selective, and controlled transfer of populations from one area to another and a (re)constitution of social life. Settlement and resettlement are not necessarily schemed, but they often are paired with the concept of a scheme. The word scheme (settlement scheme, resettlement scheme, used synonymously with project), basically refers to the idea that the (re)settlement is organized as an elaborate and systematic plan of action, including provision for shelter construction and the reconstruction of livelihood, a plan which is specified in time and scope and for the execution of which identified personnel and resources are allocated. However, as we shall see, resettlement is not necessarily schemed. In Turkey during the 1990s, hundreds of thousands of Kurds were evicted by the state from their homes and villages. This represented a resettlement, as opposed to just evacuation, insofar as the Kurds were forcibly directed from rural to urban entities, in the initial instance. They were not resettled in a strict sense of e.g. rehoused, but only in the looser sense of being told which town/city to go to first (the nearest, generally, after which they could migrate elsewhere, to other cities, which many did). This could be termed an orchestrated or tracked resettlement, as opposed to a specified or schemed resettlement.4
The root—settle, affixes—re & ment > re + settlement or resettle + ment. The territory from which the Kurds were evacuated was not, by and large, resettled. It was just left, empty. 3 4
introduction
5
The study of resettlement is biased towards the study of schemes (specified resettlement). Such schemes are attractive by virtue of their visibility and their relatively clear boundaries. Although the literature on resettlement has been growing since the 1960s and is now quite rich, it cannot really be said that an organized body of theory about [re]settlement schemes has emerged. Only a few attempts have been made to develop a comparative or analytical framework (notably, Shami 1993; Chambers 1969; Muggah 2003a, 2003b). Not only has research been constrained by scholars informed by the particular approaches and perspectives of their own disciplines (a conceptual dynamic), but it has also been dictated by the division between the resettlement of oustees and of refugees as determined respectively by the World Bank and UNHCR, which have been allocated responsibility for the categories of people created (an institutional dynamic). This division has been particularly restrictive, as the two institutions in question have created research fields that are stubbornly separated from each other and have attracted distinct sets of academic followers. It is a compartmentalization that undercuts conceptualization and undermines research (Voutira 2000). The institutional logic of resettlement studies is also largely responsible for the prevalence of engineering approaches, i.e. studies that are supposed to contribute to improvements in the design and performance of resettlement practices. These have encouraged a type of research that concerns itself with ‘practicalities’ related to resettlement, such as criteria for the identification and selection of settlers, the organization of population transfers, the provision of shelter and reconstruction of livelihood, and organization of the process in more efficient ways and at lower costs (Stola 1992). For example, although Robert Chambers’ (1969) work on settlement schemes in tropical Africa includes an abstract analysis of resettlement, it is primarily concerned with practical policy implications, including the allocation of activities, staff, and definition of roles. Another (and more recent) example is Michael Cernea’s (2000) model for the prevention of adverse effects in resettlement, developed for the World Bank. The starting point of Cernea’s work is what he refers to as a fundamental dilemma in the displacement of populations for reasons of development. The development claim is that resettlement may be engaged in to improve living conditions of the people as a whole, but such resettlement actually destroys the means of existence of those people who are displaced, often resulting in impoverishment. Such adverse effects of developmental resettlement
6
chapter one
are justified on the basis of the greater good of the greater number, a principle that Cernea dismisses as intolerable, because in reality it justifies an unequal distribution of costs and benefits, in which some people enjoy the gains of development while others have to bear its pains. Instead, his aim is to prevent the negative effects from occurring in the first place. Thus, he develops a resettlement model based on the sociological concept of risk.5 Cernea argues that all forced displacements are prone to risks, but not fatally condemned. He proposes to deconstruct risk into its so-called ‘principal components’ and target strategies to prevent their occurrence. The prediction of risks has to serve as a ‘self-destroying prophecy’. Cernea identifies eight resettlement risks: landlessness, homelessness, joblessness, food insecurity, increased morbidity, marginalization, loss of access to common property resources, and community disarticulation. Overcoming the pattern of impoverishment requires risk reversal. The directions of strategies to be taken are from landlessness to land-based settlement, from homelessness to house reconstruction, from joblessness to re-employment, from food insecurity to adequate nutrition, from increased morbidity to improved health care, from marginalization to social inclusion, from loss of access to restoration of community assets and services, from social disarticulation to networks and community rebuilding. The aim of the model is not only to foresee adverse events, but to anticipate and prevent them (the ‘self destroying prophecy’). The risk model is attractive for those ‘doing’ resettlement, and tends to turn social scientists into social engineers concerned with improving resettlement practices. The motivation for the present work, however, was to question resettlement itself. This implies that enquiries are spread beyond disciplinary limits, and across the institutional divide of ‘development’ and ‘conflict’ related resettlement. The aim is to probe state-institution philosophies about the organization of social space and the constitution of social life.
5 Cernea defines risk as the possibility and probability of an intervention triggering adverse effects within a certain course of social action.
introduction
7
Space, State and Nation A fundamental position taken here is that resettlement and settlement practices in the 20th century represent a modern phenomenon. From the outset, I would like to distinguish modernity from modernization. Originally, in the 18th century, modernity was understood as a project of liberation and radical democracy, comprising the idea that every citizen is a part of the sovereignty and can acknowledge no personal subjection (Paine, 1791–1792). During the course of the 19th and 20th centuries, however, another interpretation of modernity became dominant, one based on the idea of state-based nation (and culture) and a capitalist mode of production. A political and academic project founded upon the nationalist and capitalist usurpation of modernity emerged in the 20th century and was developed into a major theme in post-second world war social and political sciences. The concept of modernization first became fundamental in the vocabulary of American and European scholarship during the 1950s and 1960s, aiming to provide an alternative explanation of development to Marxist scholarship. More than just a research object, however, modernization was treated as a guiding principle. In the economic domain, it was associated with the development of capitalist relations of production (‘free’ labor, ‘free’ market); in the political domain, it was equated with the creation of powerful authoritarian states (with a particular role for military institutions, since the perceived impersonal social relations of the military were considered exemplary of what was universal in modern society); while in the cultural domain, it was held responsible for the production of reform, and contributed to the creation of state-based nations (such as Turkey). It was in this nationalist cultural context of modernization that resettlement emerged, as a tool of states and proto-states, a technology contributing to the realization of modernity’s nationalist content, and with a scope varying from the assimilation of peoples to the physical ‘cleansing’ and appropriation of land. Its aim is the territorial production of culture. With Michael Mann (2005: 2), I agree that, as a technology, ethnic cleansing is essentially a modern phenomenon, its death toll in the 20th century dwarfing that of all previous centuries. Yet it is not so much the product of democracy (demos), but resulting from the division of people into cultural categories, that is to say, producing and productive of nationalism. A key feature of modernization, as a project emerging from the idea of state-based nations and the production of homogenous cultures, is
8
chapter one
that is destroys in order to create (Harvey 1989: 16). This image of creative destruction—originally introduced in 1942 by the economist Joseph Schumpeter to depict the process of industrial transformation—is of crucial importance as it is itself derived from the modernist project in two different ways. At an abstract level, the project of modernity is faced with the dilemma of how to create a new world that transcends the old one. Since modernity aims at the transformation of society, a transformation that involves a transition from traditional to modern social structures and cultures, the discourse of modernity necessarily includes a destruction of the old society—and of course we have seen this manifested in countless (and accelerating) ways across the world over the last couple of centuries.6 More particularly, however, modernity was transformed from a project of liberation from within into a project imposed from above. Developed into a state-based, nation-building project, the social became the mere object of transformation (instead of subject of its own history). Modernization emerged as a narrow, sterile path from the deeds and discourses of political elites, who considered the destruction of the old society so urgent that it seemed not to matter what methods were used This idea has not be expressed any better than by the words of Turkey’s founding father, Mustafa Kemal (Atatürk), who once said in a speech that ‘civilization [i.e. modernization] is such a fire that it destroys those who ignore it’ (Kasaba 1997: 22–30). The literary archetype is Goethe’s Faust, who forces everybody to create a new (social) landscape and eventually deploys Mephistopheles to kill an old couple who live in a small cottage by the sea for no other reason than that they do not fit his master plan (Harvey 1989: 16). It is difficult to separate the project of modernization from the destruction through which it has been achieved and, moreover, the violence with which this destruction has been accomplished. The theory itself holds that violence is an anomaly, associated merely with the process of transformation. Modernization theorists argue holds that violence is rife in societies that have not yet fully modernized and are in the process of transition (from traditional societies based on mechanical
6 Key works in this respect are Eric Hobsbawm’s Laboring Men (1964) and the book he co-authored with George Rude, Captain Swing (1969). Other studies employing the idea of creative destruction are those of Walker Connor on nation building and nation destroying (Connor 1994), and Eugene Weber on turning peasants into Frenchmen (Weber 1976).
introduction
9
solidarities to modern societies based on organic solidarities). In modern society, it is argued (naively), all the members contribute in an organic matter to the total functioning of society, and social relations between individual members and between members and the collectivity are constituted in a peaceful environment; in the process of a changeover to such a society, however, when social roles and rules have not yet been clearly defined or adapted to each other, when social structures and social culture are in a process of integration and state institutions are weak, then imbalances might occur which cannot be legitimized or mediated by (state) institutions—resulting in violence. Violence is a state of exception related only to the stage of transition, a symptom of unfinished modernization (Soeters 2005: 119–121). However, the modernization project, qua project of creative destruction, itself both indirectly elicits and directly generates violence. An important way in which violence is directly generated in the creative destruction aspect of the modernization project is through the institution of the army. The role of violence in the achievement of modernity is aggravated, defined even, by the fact that, as a project to be implemented in non-western countries (the primary object of modernization-theorists), the (national) army was identified as the most important ‘actor of transformation’. The army supposedly modernizes society from above, since, according to modernization theorists, it had to be assumed to be an institution both modern and capable (equipped with the means to use force against opponents and impose its project, i.e. to construct modern society) (Koopmans 1978).7 Summarizing, not only is creative destruction and the associated violence a condition of modernization, but also the army, the most important state institution in which the monopoly on violence is institutionalized, is considered the main actor producing and reproducing modernity. Thus, violence is an intrinsic feature of modernity. This may come as something of a shocking conclusion, perhaps.
7 In Turkey, the army is very clearly cast in such a role (Knudsen 2005). From the very foundation—in armed conflict and definition through the personality of Atatürk, general and then president—to the periods of military dictatorship and institutionalization of military power through the National Security Council, MGK (Milli Güvenlik Konseyi), the army has, in varying degrees, acted, presented itself and been perceived as the guardian of Kemalism (i.e. modernization) (see also Ch. 4.4).
10
chapter one
Space Resettlement is a spatial practice. Not long ago, space was understood simply as an empty area, like a blank canvass. This idea goes back to discourses on space developed in the 17th century by René Descartes and Isaac Newton. In Descartes’ geometrical scientific method, space was conceptualized as uniform and continuous in all directions. Newton recast the idea as absolute space, which, in addition to its geometric regularity, was held to be independent from all time, matter and motion (Lefebvre 1991: 1–2; Poovey 1995: 28). Historically, twentieth century sociology inherited a conception of space from classical physics as essentially abstract, as fixed and independent of what was in it. A fundamental assertion of this work is that space is not to be viewed an abstract entity external to society, but that each society is characterized by a new organization of space and of people in space, a postulate which is directly derived from the work of Henry Lefebvre on the production of space. Lefebvre argues that each society forges or appropriates its own space, whence the need for a study of that space, with the initial requirement being that we apprehend it as such, in its genesis and form and with its particular centers. Space is a product, and our knowledge of it must be expected to reproduce and attempt to expound the process of production—as, for example, does Marshall Berman in his description of the modern landscape as socially constructed (Berman 1983: 16–19), or Kerem Öktem in his analysis of the transformation of the Anatolian settlement of Urfa into a Turkish city as an example of the construction of geographies of nationalism (Öktem 2005). The object of our interest should shift from things in space, as if space relates to us as an external object, to the actual production of space, or, space as a product (Lefebvre 1991: 31 and 36–37). In seeking to understand space as a product, Lefebvre introduces three moments of social space—the perceived, the conceived and the lived. These three moments correspond respectively to spatial practice (the externalized material environment), the representation of space (a conceptual mode, a way of thinking and understanding about space), and representational space (the affective experience of space and its associated symbols and signs). Spatial practices are comprised of physical structures, the network of roads and buildings the demarcations of land into plots, etc; representations of space are conceptualized abstract spaces, conceived as knowledge and discursively constructed by scientists, planners and engineers, etc.; and representational space is the space
introduction
11
of everyday life, which has an affective kernel, a symbolic meaning, such as a house of prayer, a coffee-house, or a shopping center. The production of space takes place through a ‘trialectical’ relation between material environments, conceptual modes of knowing, and lived experiences (Lefebvre 1991: 31). This work is informed by the practice and representational aspects of Lefebvre’s trialectic, namely, the dynamics of spatial practices (the construction of a material environment in the southeast of Turkey) and the engineering of lived experiences (such as turning Anatolian Muslim into Turks). In addition, and related to, spatial practice and representational space, two contradictory modes of representing modern space are employed. The classical understanding, influenced by such thinkers as Emile Durkheim and Ferdinand Tönnies, in which modern space is conceptualized as urban and industrial, is distinguished from an alternative understanding, in which modern space is represented by an egalitarian settlement type, beyond the rural-urban and agricultural-industrial divides, creating a hybrid rurban or agrindus form. Both of these representations of space, it will be argued, are to be related to particular spatial practices (urbanization as the construction of urban settlements and the physical transfer of people from the countryside to cities, and urbanization as the development of a new type of settlement in the countryside, forging a population neither urban nor rural). These two forms of urbanization were both identified by their proponents with the production of a homo-nationalis. The project of producing space, in the context of producing nationbased states, is essentially a project of binding political authority and cultural identity to space, or, more specifically, to the creation of the nation based state as a territorium. This attachment of a political-cultural identity to space, or the production of its spatial belonging (territorium), has been one of the central features of nationalism, and needs explanation. The production of the political-cultural spaces of nationalism is a matter of constant concern, and a process that cuts through the segmentation of Turkish history into the ‘traditional’ Ottoman era and the ‘modern’ republican era. The binding of political and cultural authority to space did not end and did not start in 1923, with the proclamation of the Republic of Turkey and recognition of its borders by major European states in the Treaty of Lausanne—although, of course, these events did first announce Turkey as a nation-state and determine its initial international borders, thereby establishing the space over which the new regime was declared sovereign and allowed to develop and implement
12
chapter one
its own understanding of national space and associated spatial practices. As will be discussed, a particular concern was the establishment of state institutions in rural areas, where custom laws prevailed, where feudal ties and traditional practices continued to hold sway and where, from the perspective of the state, an administrative vacuum existed. Concerned with engineering its national project, the new political regime in Ankara thought that the principal shortcoming of rural Anatolia was the dispersed structure and small size of the myriad rural settlements. Relatively inaccessible to the limited resources of the nascent state and with the innate resistance to change of rural settlements anywhere, these villages and hamlets were considered fortresses of resistance against the national project. Politicians and political thinkers were convinced that a new, ‘coherent’ structure of rural settlements, importantly of greater size and integrated in administrative centers, was needed in order to be able to establish the authority of the state in the countryside and develop a national body of people, the population. Box 1.1 The Borders of Turkey as Determined by the Treaty of Peace with Turkey, Signed at Lausanne, July 24, 1923 (Source: Carnegie Endowment for International Peace, 1924) Section I. I. Territorial clauses. Article 2. From the Black Sea to the Aegean the frontier of Turkey is laid down as follows: (1) With Bulgaria: From the mouth of the River Rezvaya, to the River Maritza, the point of junction of the three frontiers of Turkey, Bulgaria and Greece: the southern frontier of Bulgaria as at present demarcated; (2) With Greece: Thence to the confluence of the Arda and the Maritza: the course of the Maritza; then upstream along the Arda, up to a point on that river to be determined on the spot in the immediate neighborhood of the village of Tchorek-Keuy: the course of the Arda; thence in a south-easterly direction up to a point on the Maritza, 1 km. below Bosna-Keuy: a roughly straight line leaving in Turkish territory the village of Bosna-Keuy. The village of Tchorek-Keuy shall be assigned to Greece or to Turkey according as the majority of the population shall be found to be Greek or Turkish by the Commission for which provision is made in Article 5, the population which
introduction
13
has migrated into this village after the 11th October, 1922, not being taken into account; thence to the Aegean Sea: the course of the Maritza. Article 3. From the Mediterranean to the frontier of Persia, the frontier of Turkey is laid down as follows: (1) With Syria: The frontier described in Article 8 of the Franco-Turkish Agreement of the 20th October, 1921 (2) With Iraq: The frontier between Turkey and Iraq shall be laid down in friendly arrangement to be concluded between Turkey and Great Britain within nine months. In the event of no agreement being reached between the two Governments within the time mentioned, the dispute shall be referred to the Council of the League of Nations. The Turkish and British Governments reciprocally undertake that, pending the decision to be reached on the subject of the frontier, no military or other movement shall take place which might modify in any way the present state of the territories of which the final fate will depend upon that decision. Article 4. The frontiers described by the present Treaty are traced on the one-in-amillion maps attached to the present Treaty. In case of divergence between the text and the map, the text will prevail.
To change life, one has to change space. The credo that for the production of new social relationships in society one needs to produce a new social space was important for the new political elite in the Republic of Turkey. The generals and politicians, bureaucrats and advisors of the new regime were not only concerned with the development of a compact space as a means for the ‘technical’ downscaling of administration: they also believed that the construction of a new settlement type and a new settlement structure would contribute to the creation of new social relations. Localities were now considered instances of the nation, which was represented as everywhere the same, from west to east and north to south. But it still had to be produced. The uniform application of particular elements in the design of new villages, becoming prevalent in the 1930s, aimed to appropriate local places in the abstract space of the nation-state. In these new villages, a main street would run to a square (Republican Square) with a statue of Atatürk in the middle (to emphasize his centrality in the new social order); and at the head of the
14
chapter one
square there would be a People’s House (Halk Evi), a building designed for the teaching the ideals of the Kemalist revolution, for the party’s exercise of political power, and for the administration of the locality. The centrality and integration of the cultural, political, and administrative institutions of the state (and the manifest absence of a mosque) is revealing from the point of view of the social relations the state aimed to construct in the public domain. Space was represented as isotropic and materially associated with homologies and seriality, e.g. reproducible products (Poovey 1995: 29). The most important reproducible product became the Turkish man (and woman).8 In Turkish scholarship, resettlement is treated as one of the most important instruments of nation-state building. In general, Turkish social science literature holds to the proposition that resettlement gave Anatolia its Turkish imprint (Ari 1995; Karpat 1985) or, more specifically, contributed to a ‘nationalization’ of the petty bourgeoisie (Keyder 1979–1980). Other writers, however, have emphasized the destructive content of resettlement (Be ikçi 1991a; 1991b; Van Bruinessen 1997), highlighting how the project of modernization involves destruction in order to create (Harvey 1989: 16). The Turkish sociologist, İsmail Be ikçi, has clearly demonstrated this in the Kurdish case with a series of seven studies in which Kemalist policies towards the Kurds are analyzed as instances of ‘nation-destruction’. Two of the seven volumes of this important work are concerned with the (re)settlement of Kurds. Volume 1 discusses the Settlement Act of 1934, which constituted the legal framework for the resettlement of Kurds as a method of assimilation; and in Volume 4 there is discussion of a special law passed in 1935, which placed the province of Tunceli under military rule and prepared the way for brutal military campaigns in 1937 and 1938, in the course of which a considerable part of the population was killed
Female emancipation was also a central plank of the Kemalist project (e.g. equality of inheritance was enshrined in law just three years after the foundation of the republic, and complete women’s suffrage was gained as early as 1934)—the state rejection of traditional, Islamic-influenced gender inequalities being intricately interwoven with the production of a national culture as one aspect of its secularist, post-Ottoman modernism. Indeed, from the outset, the Turkish nation-state was self-consciously defined in terms of modernity. Other aspects productive of a national culture of modernism, in addition to the central focus here on spatial practice, included Mustafa Kemal’s ‘reforms’, which ranged from abolishing the caliphate to purifying the Turkish language of foreign words, from adopting a code of civil law to replacing Friday with Sunday as the official day-off, from developing a unified system of universal education to banning the fez. 8
introduction
15
and many of the survivors deported to western Turkey (Bruinessen 2003–2004). Compared to the discussion of resettlement, settlement has been seriously understudied in Turkish scholarship. The spectacle of deportation has attracted considerably more attention than settlement policies, for obvious reasons, but actually it was the latter which entailed the more comprehensive attempt to assimilate people and appropriate land, and by a long way. An exception to the Turkish academic neglect of settlement is Fuat Dündar’s study of the settlement policy of the Committee of Union and Progress between 1913 and 1918 (Dündar 2002). Dündar analyzes the immigration and settlement policies of Turkish nationalists during the last decade of the Ottoman Empire, interpreting them as instances of ‘population politics’. In this work, a development of Settlement Wars (2006), my doctorate thesis, settlement and resettlement are discussed as means by which the Turkish state has created and managed its territory, through both the (integrative) production and (dislodging) productive destruction/destructive production of (the) population(s). The State Resettlement and settlement practices will be discussed as modern methods of nation- and state-building, implemented by violent means if necessary. The concepts of state, nation and nation-state need some proper introduction, since a confusing conflation of key concepts (‘a nation is a state is a people is a . . . .’) has characterized the study of states and nations (Connor 1994). We cannot, like Humpty Dumpty, the curious figure in Lewis Caroll’s world, just make words mean what we choose them to mean. We need to be clear about the definitions we adhere to, while at the same time recognizing the malleable character of concepts. Concepts do not have fixed meaning, independent of space, time and the perspective of the individual employing them. In a famous definition, Max Weber says that a state ‘is a human community that successfully claims the monopoly of the legitimate use of physical force within a given territory’ (cited in Faulks 1999: 20). This definition is problematic for two reasons. First, the juxtaposition of ‘physical force’ and ‘legitimate use’ is contradictory. If the State is legitimate, why does it have to rely per definition on violence to control its population? Second, the definition of the state as a ‘human community’ leads to the inter-changeability of the concepts state and nation.
16
chapter one
A better definition is given by Anthony Giddens (1989: 301), who says that ‘a state exists where there is a political apparatus [. . .] ruling over a given territory, whose authority is backed by a legal system, and by the capacity to use force to implement its policies’. What Giddens calls a ‘political apparatus’ is specified by Faulks’ definition (1999: 2), as a set of tightly connected institutions concerned with the administration and control of a geographically determined population. Giddens and Faulks are to be preferred to Weber here for two reasons. First, their definitions do not place the emphasis on the use of violence—but on the administration and control of a geographically determined population (backed by force). Second, the state is not equated to a community, but conceptualized as a set of interconnected institutions which prevails within a particular territory. However, we should also be precise about the description of the state as a political apparatus (Giddens) or set of institutions (Faulks). In his “Notes on the Difficulty of Studying the State” Philip Abrams (2006: 118) argues that the state does not exist as a real entity and at the best is an abstract-formal object. Alternatively, following the Marxist political scientist Ralph Milliband, he proposes to speak about a state-system as a cluster of institutions which does not have any particular unity (ibid.: 119, 125). The idea that we should abandon the idea of the state as a freestanding entity—be it agent, structure, instrument or organization— is also shared by other social and political scientists (Mitchell 2006: 184). The idea of the state as a cluster of institutions, not necessarily ‘tightly connected’ and in particular circumstances ‘loosely related’ or even9 ‘contradictory’ (a cluster of institutions, that is, concerned with the administration and control of a geographically determined population), will prove to be an important notion in the analysis developed in this work. It will be used to explain contradictory policies and practices in settlement and resettlement by various state institutions. Regarding the relationship between state and nation, Wallerstein (2002) holds that a systematic history of the modern world shows that, in almost every case, states have preceded nations. Wallerstein further argues that any particular state creates a corresponding nation via a functionalist logic: states find themselves frequently threatened by In the modern world, culture does not mark status, argues Gellner, but the boundaries of political units. Here, he follows the footsteps of Max Weber, who strongly emphasizes the impact of bureaucratization on culture. For Gellner, the coincidence of state and culture took us to the age of nationalism (Gellner 1983). 9
introduction
17
internal disintegration and external aggression, against which the idea of ‘the nation’ provides necessary cohesion.10 The historian Hobsbawm also argues that it is not nations that produce states, but vice verse (Hobsbawn 1990). Yet he does not consider the nation to be inevitable or a functionalist artifact of statehood, as Wallerstein does, but sees it as a product of a political phenomenon called nationalism that entered human history at the end of the 18th century. In itself, the definition of nationalism is not very problematic. Hobsbawm defined nationalism as the principle holding that the borders of state and nation should coincide (Hobsbawn 1990: 9). Likewise, Gellner defines nationalism as the belief or principle that cultural and political units in society should be congruent (Gellner 1997: 240). However, the principle of congruency is based on the widespread assumption that a nation is to be considered a ‘[human] collectivity existing within a clearly demarcated territory’, even though the idea that human collectivities are spatially organized in separate entities is extremely questionable (Giddens 1985: 4). Nations do not exist as packaged bundles of people waiting for a state to be drawn around them (Taylor 1985)—on the contrary, the territorial organization of (imagined) human collectivities is a main concern of ‘nation-states’. Nations do not explain states, but need explaining themselves (Öktem 2004). Nations and Nationalism Studies of nations and nationalism divide roughly into two streams, which we may categorize as the ‘primordialist’ and ‘modernist’. The primordialist holds that nations are perennial and natural. Acceptance of the latter entails acceptance of the former, but not the other way around. It is possible to argue that nations and nationalist ideas and sentiments have existed throughout history, without being ‘in the blood’. Perrenialists may simply claim that the nation is a larger and upgraded kind of ‘ethnicity’,11 as Anthony Smith does (Smith 1986: 12–13). 10 One has to be careful not assume that ‘nation-building’ automatically produces cohesion. Jean-Paul Sartre rightly argues that nation building also produces ‘difference’, e.g. minority ethnic, religious, etc. groups. This is also elaborated upon in ‘Nation-Building or Nation-Destroying’, an article written by the political scientist Walker Connor (1994), which discusses some of the difficulties nation-states have with ‘national minorities’, and the resistance of national minorities to integration and assimilation, a collective resistance that may contribute to the development of a collective (national) identity. 11 The antique term ethnicity (ethnos) covers a variety of usages, but the common
18
chapter one
According to Smith (ibid.: 22–32) ethnos has six dimensions: a collective name, a common myth of descent, a shared history, a distinctive shared culture, an association with a specific territory, and a sense of solidarity. The nation originates from the ethnos, according to Smith, and the main difference between the two is that the latter does not have a common polity. The modernist approach to nations and nationalism holds that they are essentially modern phenomena. The modernist Ernest Gellner (1964; 1983; 1997) asserts that i) nationalism is the inevitable consequence of industrial society (Gellner 1983: 39; 1997: 240), and ii) nationalism engenders nations, and not the other way around (Gellner 1964: 55). Underlining nationalism’s invented character, Gellner further argues that nationalism is selective in the use it makes of pre-existing cultures, reviving dead languages and traditions, and restoring fictitious, pristine purities. ‘The cultural shreds and patches used by nationalism are often arbitrary historical inventions. An old shred and patch would have served as well’ (Gellner 1983: 56). Gellner rejects the idea that the nation is an up-dated version of the ethnos, holding it to be scarcely conceivable that nations would have emerged had ethnic miniorganization rigidly remained strong everywhere, and suggests instead that the emergence of nations depends on the abolition of ethnos, with two empirical exceptions to this rule, the Somalis and the Kurds (Gellner 1983: 85–7). Gellner’s approach to nationalism is criticized for being infused with functionalism (the idea that social phenomena can best be explained in terms of the functions they perform in sustaining society) and determinism (the idea that phenomena are governed by causal and universal laws). He is a functionalist insofar as his analysis of nationalism is limited (reduced) to the contribution that nationalism makes to the functioning of industrial society. Nationalism and then nations are established because of the social differentiation attendant upon the occupational mobility of the industrial economy and the consequent need for a new integrating culture: this integrating culture itself then requires the state for protection, and the state needs the nation for denominator seems to be the sense of a number of people living and acting together. The concept of ethnos is not equal to kinship ( genos). The main difference between ethnos and genos is that the first is more strictly socio-cultural and the second has a sociobiological component, links through common descent (Smith 1986).
introduction
19
legitimacy and thus to brand its flock with a homogenous culture. Gellner is a determinist in so far as he argues that nations and nationalism are an inevitable consequence of unfolding modernity. However, in his attempt to demystify nationalism and to show that nationalism is not only an inevitable but also a rational product of modernity, Gellner does appear to have missed out perhaps the most important of all 20th century facts, World War II and the Holocaust, the direct results of nationalism as expressed through the rise of fascism and its culmination in Nazism. In his concern to explain nationalism as a rational offspring of modernity, Gellner only discusses (apparently) liberal expressions of nationalism, completely ignoring its brown form which unleashed both a global military conflict and a genocide, each unparalleled in human history (Anderson 1994: 204–5). For a thorough understanding of nationalism and the establishment of nations, we cannot fail to notice the dimension of shared meaning or collective representation—i.e. the notion of identity fulfilled. This is the basic premise in Benedict Anderson’s ‘Imagined Communities’, a corrective to the reductive modernist account. Influenced by post-modernist encounters with issues of identity, Anderson defines the nation as a territorially limited imagined sovereign community (Anderson 1991: 5–7). Nations are imagined because the members of even the smallest nation will never know all or most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of a community. The nation’s members feel connected to people whose existence they do not know, except in abstract numerical terms (‘We are a nation of 67 million people’). Nations are imagined as territorially limited because even the largest and most irredentist nation has finite boundaries beyond which lie other nations. Unlike Christianity or Islam, nationalists do not dream of a day when all people in the world join their community. Nations are imagined as a community because, regardless of actual inequality, exploitation, and submission, the nation is conceived in terms of comradeship. Finally, the nation is imagined as sovereign because the supreme authority is located in the nation itself. Most social scientists have great difficulty in giving a definition of the nation. This is vividly articulated by Eric Hobsbawn (1990), who challenges us to imagine that an intergalactic historian has landed on our planet after so-called smart nuclear weaponry destroyed human life but left libraries and archives intact. The extraterrestrial finds out that to understand social life over the last two centuries some knowledge of
20
chapter one
the term nation is necessary. The visitor sees that the idea of the nation and use of the term was everywhere—for example in politics (nation-state, national elections, etc.), in culture (national literature, national identity, etc.), and in geography (national rivers, national boundaries, etc.). Most likely the extraterrestrial would also learn that, as Walter Bagehot put it, people knew what the nation was when you do not ask, but could not very quickly explain or define it (Hobsbawn 1990: 1). The problem of definition is aggravated by careless use of language. The term nation is frequently misused, both in common language and in social sciences (Connor 1994: 92–93). For example, official expressions such as the League of Nations, United Nations are hopelessly confusing the term nation with that of state. In scholarly approaches to the nation there has also been a tendency to collapse the difference of meaning between these two (even in the naming of the academic subject, international relations). Such approaches reflect an idea, traceable to the eighteenth century, that there is no other way to define nation other than as a state, an assumption captured in Karl Deutsch’s famous definition of a nation as ‘a people who have hold of a state’ (Keane 1998: 85). Historically, the word nation is derived from the past participle of the Latin verb nasci meaning ‘to be born’ and seems to imply ‘common ancestry’ and ‘place of origin’ (c.f native land, native people, etc.) (Connor 1994: 94). The initial concept was used in a pejorative sense, used in classical Rome to designate foreigners from a particular region whose status, as foreigners, was below that of the citizens (cf. colonial reference to the natives). At the University of Paris during the Mediaeval period, the word nation and became used in a neutral sense to designate communities of students, referring to people united by place of origin. Four nations were differentiated: l’hontable nation des France (students from today’s France, Italy, and Spain), la venerable nation de Normandie (Holland and Normandy), la fidele nation de Picardie (the northern French departments on the English Channel, excepting Normandy), and la constante nation de Germanie (England and Germany). The communities of students served as support groups and the word came to refer not only to the idea of communities of origin but also to communities of opinion and purpose. By the late 13th century, the concept of the nation acquired yet another meaning—that of an elite (Greenfield 1992: 3–9). With the decline of the Carolingian Empire, a new sense of collective identity—national awareness—began to emerge as a strong social force. The nobility and clergy championed this process of nation build-
introduction
21
ing. They used the term natio to highlight their sharing of a common language and common historical experiences. The natio did not refer to the whole population of a region, but only to those classes that had developed a sense of identity based upon language and history, and had begun to act upon it. The natio in this cultural sense was seen as a distinctive product of its own peculiar history. From the fifteenth century onwards, the term nation was employed increasingly for political purposes. According to the classic definition of Diderot, a nation described a people who shared common laws and the political institutions of a given territory. The political conception of a nation defined and included civil society, i.e. those citizens entitled to participate in politics and to share the exercise of sovereignty—and had fundamental implications for the process of state building. Struggles for participation in the state assumed the form of confrontations between the monarch and the privileged classes, whose interests were often represented by a parliament. These classes considered themselves the advocates of ‘the nation’, in the political sense of the term. In opposition to the monarch, they insisted they were the defenders of ‘national liberties’ and ‘national rights’. During the period of the Enlightenment the struggle for national identity was broadened and deepened to include the unprivileged (or less privileged) classes. Eventually the nation came to include everyone. In the course of the 19th and 20th century, however, the concept altered once again—it acquired a cultural meaning, referring to a unique people. In its history, therefore, the concept of the nation has undergone four transformations, changing from a group of foreigners to a community of opinion, then to an elite, to a sovereign people, and finally to a unique people (Keane 1998). The modern idea of the nation carries connotations of a community shaped by common descent, culture, language, aspirations, and history. It provides a particular form of collective identity in which people, despite their routine lack of physical contact, consider themselves bound together because they speak the same language; because they inhabit a defined territory, are closely familiar and have an affective relationship with the local ecosystem; because they share customs, including internalized memories of an historical past, itself consequently experienced in the present tense as a pride in the nation’s achievements and an obligation to feel ashamed of its failings (Keane 1998: 86). Clearly, the idea of (a unique) people as a nation has multiple dimensions. It is
22
chapter one
not simply an ethnic community, or cultural community, or linguistic community, but a complex mix, with the relevant importance of each of these constituent elements liable to change. The most important figure in the creation of a Turkish nationalism was Ziya Gökalp. Born in 1875/6 in the provincial Ottoman city of Diyarbakır (ironically, perhaps, in the light of that city’s place in TurkishKurdish geography and history), and dying in 1924, shortly after the establishment of Turkey, Gökalp had tended to be a loner in Turkish nationalist circles during his lifetime (Deringil 1998: 65). Nevertheless, his work came to occupy a position of prominence, formative in the development of Kemalism and instrumental to Atatürk’s reforms. Gökalp’s approach centered upon three processes: Turkification, Islamization and Modernization (Türkle mek, İslamla mak, Muasırla mak, the title of his 1918 nationalist manifesto) (Gökalp 2004). The aim of Turkification (Bozarslan rightly emphasizes that the suffix -tion indicates that Gökalp considered it a process rather than a given fact) was the socialization of people into the Turkish culture (Bozarslan 1996: 137), i.e. to speak Turkish and to believe in the ideal of the Turkish nation. Gökalp is said to have been a convinced secularist in the sense that he favored a strict separation of state and religion (Zürcher 2002a), but he was also convinced that Islam constituted the soul of Turkish society, thus distinguishing the Turkish nation from those in the West (similar concerns regarding the relationship of Europe to modernization led him to couple Turkey with Japan as the other developing state forging a new path of progress). Modernization (signifying material westernization, i.e. economic development, technology and science) was supposed to enable the Turkish nation to catch up with Europe and attain the most developed stage of civilization (Bozarslan 1996: 137). Gökalp’s synthesis of Turkification, Islamization and Modernization has marked the official state policy, except for a period of radical secularism in the 1920s and 1930s, when he apparently fell out of favor (and his sonin-law, Ali Nüzhet Göksel, even had considerable difficulties getting his work re-published) (Göksel 1955). For Gökalp the nation was a cultural entity, attached to space (Gökalp 1922). However, a fundamental ambiguity in this cultural definition of the nation—one Gökalp does not explicitly deal with—is that changing one’s culture is almost impossible, since it is also a product of history (Zürcher 2000). Following this theme, the Turkish nationalist Nihal Atsız argued that nations are the product of thousands years of history and that no one can adopt another culture within a couple of
introduction
23
years or even centuries. Atsız concluded from this that the Kurds had to leave Turkey and ask the United Nations to allocate them a place to live somewhere in Africa (darkly, he advised the Kurds to ask and learn from the Armenians what might happen to them if they did not). Such ideas were no oddity. A Turkish Minister of Foreign Affairs, Tevfik Rü tü, had argued in 1927 in the presence of the British ambassador to Turkey, Sir George R. Clerk, that ‘the Kurds (. . .) are inevitably doomed.’ ‘Their cultural level is so low’, he continued, ‘their mentality so backward (. . .), they will die out, economically unfitted for the struggle for life in competition with the more advanced and cultured Turks, who will be settled in Kurdish districts’ (Simsir 1975: 98). Thoughts like these were not reserved for the Kurds alone. Atsız also proposed to deport Gypsies to India, and, if this was not feasible, to settle them in the eastern province of Hakkari (implying, it appears, that Hakkari was a kind of ‘second class’ Turkey). Atsız also considered Circassians, Albanians, and the Laz to be ‘foreigners’. At the heart of these extremist ideas was (and is) the belief that culture is a something ‘biological’ (the primordialist, natural view), and that citizen rights should be given to ‘real’ Turks only, a conviction which has been a strong current in Turkish nationalism (Bakırezer 2004). In this context, Martin van Bruinessen also refers to the opposition to a civil definition of Turkishness in a 1981 draft of Turkey’s new, revised constitution.12 General İhsan Göksel, member of both the National Security Council (MGK—Milli Güvenlik Konseyi—at the time the defacto highest state institution in Turkey), and also of the committee set up to advise on a new constitution (Anayasa Komisyonu), stated that: ‘By accepting a person as a citizen we cannot change the blood in his veins and replace it with Turkish blood, we cannot tale the values in his heart and mind and instead rebuild him physically and spiritually with Turkish culture, Turkish virtues, the rich history of Turks and, if you wish, the racial superiority of the Turkish race’ (Bruinessen 1997: 3). However, the dominant view was that assimilation was possible and desirable.
12 Reported in the monthly review of the Prime Minister’s Office Press and Information Directorate (Office of the Prime Minister 2006).
24
chapter one The Area under Study
Founded in 1923 from the (primarily) Anatolian remnants of the Ottoman Empire, Turkey lies on the southern shores of the Black Sea, between Bulgaria and Georgia, and on the eastern and northeastern shores of the Aegean and Mediterranean Seas, between Greece and Syria. In geographical terms, Anatolia is a high plateau with several mountain ranges. Ruled by the state of Turkey, the territorial object of this study is also referred to as (part of ) Kurdistan, a strategically located region in the Middle East comprising parts of Turkey, Iraq, Iran and Syria. Specifically, we focus here on Diyarbakır, the province in which the field research was carried out. The province of Diyarbakır today covers a area of 15,355 square kilometers (in the Ottoman Empire the province of that same name was much larger, including also large parts of the contemporary provinces Mardin and Batman). Agriculture provides the main economy, with dry farming and fallow, the people mainly dependent on crop production and stockbreeding, along with petty commodity trade. Diyarbakır is located in a transition zone connecting the mountainous northern areas to the plains of Mesopotamia. It is presently the centre of the highway network connecting other important settlements in the region, such as Adiyaman, Antep, Batman, Elazig, Mardin, Urfa and Van. The railway reached the city of Diyarbakır in 1935 (‘Diyarbakır’ names both the province and its provincial capital), and was later extended outside the province to Kurtalan. Diyarbakır has air connections to İstanbul and Ankara. The population of the province is approximately 1.3 million, living in the city of Diyarbakır, 13 towns (which also give their names to the districts of which they are the administrative centers: Bismil, Çermik, Çınar, Çüngü , Dicle, Eğil, Ergani, Hani, Hazro, Kocaköy, Kulp, Lice and Silvan), along with a further 741 villages and 1,254 hamlets. This amount to 14 urban and 1995 rural settlements, indicating a rural and highly dispersed and fragmented settlement structure. Diyarbakır has been territorially mapped in different ways in various political discourses. It was not considered by the Ottomans to have any particular cultural identity, but a multiple one. Turkish nationalists, gaining power at the turn of the 20th century, conceived of the area as a part of a greater Turkish nation. Kurdish nationalists, however, mapped Diyarbakır as a part of a greater Kurdish nation, and rebelled against the spatial practice of establishing a Turkish cultural identity on the land where they lived. In 1925, Diyarbakır province
introduction
25
was the site of what was later named the Sheikh Said Rebellion (for a detailed study of the rebellion, see Olson 1989). In an attempt to sidestep these conflicting ways of representing the space (by Turkish and Kurdish nationalists), the rebellion was depicted as a religious and tribal reaction to the modernity of the republic. Likewise, the Kurdistan Workers Party, PKK (Partiya Karkerên Kurdistan), which had held its founding ‘congress’ (a meeting of 25 people) in a remote village in a district north of Diyarbakır city in 1978, is also often depicted in Turkish historiography as a product of the backward social relations allegedly prevalent in the southeast of Turkey. In the imagination of many Kurds, the Sheikh Said Rebellion and PKK uprisings are part of their political and national struggle. The quantification of the war waged by the PKK as the 29th rebellion (Kahraman 2003: 19) is an attempt to give continuity to this struggle (and an impression of a process of continuous uprising, although actually four major uprisings have taken place, at Dersim and Ağri, in addition to Sheikh Said and the recent PKK-led uprising). It is not clear that this perspective holds only for the Kurdish side. The fact that the PKK leader, Abdullah Öcalan, captured in Kenya on February 15th (1999), was sentenced to death by a Turkish court on June 29th, the very same date as the execution of Sheikh Said 74 years earlier, contributes to the idea that the uprisings all mark crisis points in a single struggle for territory between the forces of competing (Kurdish and Turkish) nationalism. Tribes Notwithstanding the struggle on behalf of the Kurdish nation, tribal affiliation continues to exert a strong social force. It proved to be an important point of reference in two of the four villages studied here, and for that reason a few words should be said about the tribe as an institution. The Kurdish tribe is a socio-political, territorial and economic unit based on descent and kinship, real or imagined (Bruinessen 1978: 40). Most Kurdish tribes have been sedentary for a long time, with villages bearing their names. Although not all Kurds are tribal, and in some areas non-tribal Kurds form the majority of the population, the tribe has a key role in Kurdish political, social, and economic life. In Diyarbakır province, the local Kurdish population uses the terms esîret, qebîle, malbat and mal to refer to a tribe and its subdivisions. In this subdivision of descending order, the mal is a patrilineage, a consanguineous
26
chapter one
male kin group each of whom is related to a common ancestor. The mal is the most important social unit, which may be composed of a nuclear family (husband, wife(s), and children), but may also contain several generations. The malbat also refers to a patrilineage, but in a broader sense, comprising several mal. The difference between mal and malbat is that the mal forms a direct line of descent from a close ancestor (mostly a grandfather), while the malbat forms a direct line of descent from a more distant ancestor. The qebîle also refers to a patrilineage, but wider again, comprising several malbat. It ought to be emphasized that not everyone differentiates between malbat and qebîle. The esîret comprises various qebîle, and, in contrast to the qebîle, in which common ancestry can be traced, common ancestry in the esîret is difficult to trace, or imagined. Mark Sykes (1915: 575), who tried to trace the common ancestry of the Milan tribe, relates how he sometimes became desperate as Milan Kurds would suddenly grow vague or change the subject while telling fragments of their history and ancestry. Legend has it that the Kurds were divided into two branches, the Milan and the Zilan. The Milan live pre-dominantly in the southwest of Turkish Kurdistan (although there are also Milan living on both sides of today’s border between Turkey and Iran), while the Zilan live pre-dominantly in the northeast (although there are Zilan in Batman too, a province in the north-west). It is said that the Milan came from Arabia and the Zilan from ‘the East’. A variant on this legend says that the Kurds were divided into three branches, the third being the Baba Kurdi, whose patron saint is Khalid-ibn-Walid, whom they hold in great reverence, saying that he converted them from Paganism to Islam (Sykes 1915: 555). Little is known about the Baba Kurdi, and it is hard to find references to them in the literature. It is suggested that they speak the Sorani dialect or are from Persian origin, but the tribes which are referred to as Baba Kurdi (Ertu i, Pinyani, Zibar and Barzan) actually speak the Kurmanci dialect of Milan and Zilan Kurds. Two of the four villages where the research reported here was carried out were inhabited by tribes that are part of the Milan, one by Xedrîkan (Xedrîk) and the other by Metîn (Metînan). Because the Milan are rather heterogeneous (with component parts from different religious and ethnic backgrounds), they are referred to as confederacy of tribes (Kiran 2003; Sykes 1915). According to legend, the Milan was once a powerful tribe, but scattered by God, who was displeased with them. Some of their component parts vanished, others remained. Ziya Gökalp counted nineteen Milam qebîle or esîret, Mark Sykes more than
introduction
27
80, including the Berazieh (Berazî), which his information held to be a confederation composed of twelve tribes (Sykes 1915). Oral tradition narrates that after their dispersal the Milan were re-established by seven tribes (the Cebikan, Cimikan, Hacikan, Kiran, Kumnaksan, Sinkan, and Xedrikan). Under the leadership of İbrahim Pa a (today also referred to by Kurds as Milli İbrahim, see for example Kiran 2003) the Milan gained considerable power at the end of the 19th century, but after his death in 1909 the confederation fell apart. The dissolving of the Milan is not to be interpreted as evidence for a gradual dissolving or extinction of tribes in general. Kurdish tribes have shown great resilience. (for a lengthy discussion of this subject see Bruinessen 2003). State policies towards tribes have not been particularly consistent. Turkey has been involved in the deportation and dispersal of tribes and their leaders in order to destroy this social formation, but, conversely, the politics of domination also strengthened the tribe as an institution. First, Turkish political parties used tribal leaders as ‘bulk vote generators’ (Barkey and Fuller 1998: 77). Participation in the political system (a district system, which allocates to each province a number of seats for the election of parliamentarians) gave tribal leaders access to national power and thus to resources to strengthen their tribal structures (Bozarslan 1996: 141–142). When their party was in power, tribal leaders had the opportunity to reward loyal supporters in the form of infrastructural investments and government contracts (Bruinessen 2003). Deputies also spent part of their time (or their assistants’ time) on individual matters, such as support in the transfer (of civil servants, teachers, and others in government service) to a new post or a similar post in another location (e.g. town instead of village), which could be a rather bureaucratic and lengthy procedure without such help. Second, in addition to its use of tribal leaders, the State also mobilized tribes and organized them as paramilitary militia to fight guerrilla movements. This gave the tribes that participated, and their leaders, access to weapons, and the right to use them, as the state shared (at least part of ) its monopoly on violence with these tribes. Villages, Lineages and Households This study focuses on villages for its case studies, a methodological choice. ‘The village’ is the traditionally privileged object of study in the field of anthropology, whereas in conventional rural sociology and rural economy it is households and farms that take primacy. The decision
28
chapter one
to take villages as units of study here was not inspired by disciplinary considerations, but determined by the fact that settlement policies were executed at the level of rural settlements. Although the village is the unit of case study, however, the main social unit in the rural settlements is not the village as a whole, but the mal. As mentioned, the mal is a patrilineage.13 It is a patriarchal and hierarchical structure organized around a male head, under whose name the mal is known. Mala Neyo, for example, is the patrilineage headed by a man called Neyo. Even after Neyo’s death, his descendants may continue to be referred to by this name if Neyo was respected and his family stays together. The male members of the family make up the core of the mal and remain lifetime members of their natal mal if they do not form a mal of their own. Women usually do not form their own mal. They are part of the mal of their father (or an older brother) until marriage, after which they become part of the (typically newly formed) mal of their husband. In this system of male differentiated mals, an exchange of women in marriage creates articulation and interconnection. To articulate mal and establish interconnections two mals can decide to perform berdêlî, a form of (double) marriage exchange in which a man from one of the mals marries a woman from the other, and the woman’s brother marries the man’s sister (so one couple will be in one mal and the other couple in the other mal). A mal can develop an extended settlement pattern, in this research implying the development of a pattern of settlement divided over both rural and urban settlement units. It appeared that a mal may form one household in the village, while splitting and spreading over several households in the town or city. The mal is the smallest social unit in villages, and may take the form of a household. It is a dynamic unit and changes in size and composition with the life-cycle of the family, a principle also described by Chayanov (Chayanov 1966). A mal may run through three different phases: first the nuclear family (husband and wife) without children, then the nuclear family with children, and then the extended family (nuclear family with children, along with parents, brothers and sisters of the husband). The fact that a mal is an economic unit implies that it has shared resources, but it does not necessary share all resources. Male members of a mal
13 The word mal is also used as a synonym for tribe, so Mala Milan is the Milan tribe.
introduction
29
may decide to cultivate the land they are entitled to together and share the revenues, while income generated from other work (wage labor in the city or income from entrepreneurial activities) is not shared. The possibility and probability of forming a mal is dependent on a man’s ability to mobilize, maintain, and hold resources, both financial and social. Where the mal owns land, the mobilization of resources is partly dependent on the hold over the land of the male head of a new mal which is separating from an existing mal, i.e. on the relative power of the man leaving his family (natal) home and establishing his own (marital home). This phase of separation and formation may be accompanied with bitter and even violent conflict. In one of the study villages, the head of an existing mal clashed with his eldest son, who claimed more and better land from his father on which to expand and maintain his mal. The conflict escalated when the son started to threaten his father and damaged his possessions (he set fire to a tractor) in order to strengthen his claim, and only came to an end with the sudden and violent death of the son. A Note on Names A pressing question during the preparation of this work concerned nomenclature: what should I call the broad region under study and in which the case-study villages lay? Naming is not a neutral activity but loaded with meaning, an important part of the processes comprising the political appropriation of spaces. I preferred to avoid the term Kurdish area(s) as it immediately attaches a specific social identity to a particular geographic region or territory. Given that identities are not exclusively and neatly arranged in space, the specification of areas as, for example, Turkish, Kurdish, Armenian, Circassian or Laz therefore constitutes nationalist appropriation, implies patterns of domination and subjugation, and is potentially dangerous (inviting, as it could, the ethnic cleansing of those who do not fit or comply with the cultural identity attached to a particular region). Therefore, the term Kurdish area(s) was not an option. The areas inhabited by Kurds have also been referred to as the East (until the 1980s) or the Southeast (since the 1980s). Both these terms comprise an area which covers two administrative regions, one of which is called the East Anatolia Region and the other the Southeast Anatolia Region, referred to as the East and Southeast for short. So the terms the East and the Southeast refer to both human geography and administrative
30
chapter one
division, which of course are rather different. Furthermore, the area is only in the east or southeast from a Turkish perspective: from a Kurdish perspective the area is in the north/northwest, located in their political imagination in the north/northwest of Kurdistan, and Kurdish nationalists are inclined to call the region North-West Kurdistan. As a euphemism for Kurdistan, local Kurds also use the term the region (bölge) or our region (bizim, bölgemiz), the latter distinguished from their region (onların, bölgeleri), which is of course Turkey (minus our region/bölgemiz). For someone doing comparative research in different parts of Kurdistan (in Turkey, Iraq, Iran, and Syria) the use of North, East, South, and West Kurdistan might be convenient. But in this research we deal with settlement issues and the dynamics of people and territory in the context of the making of a Turkish nation-state. Therefore I decided to use as terms the Southeast and the Kurdistan region in Turkey.14 (In order to prevent confusion, the formal names will be used, Southeast Anatolia Region and East Anatolia Region, when referring to the administrative entities.) The issue of naming also reared its head in another way during my research: in regard to the naming of settlements. Like most settlements in the Southeast, the four settlements I concentrated on for case studies had two names, which were referred to in daily speech as the ‘old’ name and the ‘new’ name. The use of quotation marks here indicates that the adjectives old and new do not refer to a time difference (only)—the adjective old is a euphemistic reference to Kurdish, Armenian and other non-Turkish place names, reflecting the (former) multiple spatial practices. The adjective new refers to the Turkish name, or renaming. (Re)naming places is part of a process of drawing boundaries and attaching social meaning and identity, and serves as a repository of values. It is intimately involved in the conversion of places to territories. The users of name variants such as Yerushalayim (Hebrew) and alQuds (Arab), or St. Petersburg (Monarchist) and Leningrad (Marxist), or Tunceli (Turkish) and Dersim (Kurdish) are locating these cities in different territorial maps and political agendas. When place name variants express different meanings and invoke conflicting political imaginations, 14 Of course, both names will sit uneasily with one of the parties to the TurkishKurdish conflict, the Southeast for the reasons referred to (it assumes the Turkish perspective), and the Kurdistan region in Turkey because any reference to Kurdistan by name is anathematic to Turkish nationalism and the Turkish state (e.g. in the recent decensorship of the 1982 Cannes Palm d’Or prize-winning film Yol, still the shot of a sign announcing the border of Kurdistan had to be edited out before general distribution was permitted).
introduction
31
they may evoke powerful emotions.15 It has been argued that sweeping changes in place names is inextricably linked to the ideological upheavals of nation-building and state formation. Such toponymic strategies aim to construct a new relationship between culture and space, and either subordinate or annihilate the other from topographic representations (Cohen and Kliot 1992; Öktem 2005). In Turkey, the state has systematically tried to efface ‘old’ Kurdish, Armenian, Greek and other village names, and to inscribe Turkish a meaning, on the surface at least, by replacing the ‘old’ names with the ‘new’ Turkish ones. Announcements to change ‘foreign’ village names had already been made prior to the establishment of the republic, in 1915, under the nationalist government of the (‘Young Turk’) Committee of Union and Progress—a plan that failed to materialize with collapse of the government in 1918, when Mehmed Vahdettin ascended to the sultanate (as Mehmed VI). Again, in the first three decades of the Republic the change of village names was on the agenda, as part of a general policy to erase non-Turkish identities. Ironically, waves of place name changing were initiated or occurred under the so-called liberal governments, notably the period of the Democrat Party of Menderes in the 1950s and the Motherland Party of Özal in the 1980s, each voted into power following repressive cultural politics (the latter after a period of military dictatorship). It was in 1956, during the fourth of Menderes’ five terms of office spanning the decade 1950–60, that the ‘The Special Commission for the Change of Names’ (Ad Deği tirme Ihtisas Komisyonu) was established under the auspices of the Ministry of the Interior. The Commission brought together representatives of the General Command of the Armed Forces, the Defense Ministry, the 15 In Turkey, a court case arose over the naming of streets in the city of Batman. The municipality, headed by HADEP, had selected street names as Mahatma Gandhi (anti-colonial and non-violent activist, resisting British dominance over India), Ömer Muhtar (leader of the anti-colonial movement in Libya, resisting French dominance), Halabja (a Kurdish town in Iraq, subject in 1988 to a poison gas attack by state forces under Saddam Hussein), Yılmaz Güney (director of the film Yol, a socialist and a Kurd), Mehmet Sincar (Kurdish member of parliament, killed by unknown assailants in Batman in 1993), Ahmed Arif (Kurd and poet), Zilan (the name of a tribe in the area, but also the code name of a PKK militant who committed a suicidal action killing several members of the Turkish armed forces in Tunceli/Dersim on the 30th of June, 1996), Elmedina (the Kurdish name of a village flooded in 1927), and Lale (a village in Northern Iraq, site of the holiest center of the Yezidi religion). The names allegedly encouraged rebellion against the state (e.g. Mahatma Gandhi and Halabja) and separatism (e.g. Yılmaz Güney), or belong to a ‘foreign’ (i.e. Kurdish) language (e.g. Lale ).
32
chapter one
Education Ministry, the Faculty of Letters, History and Geography of the University of Ankara and the Turkish Language Foundation. The Commission embarked on the task of the ‘Turkification’ of place names, and in 1968 approximately 30 percent of the names of the 45,000 villages that had been counted in Turkey were changed. In 1973 the Commission commenced work on smaller scale maps, changing the names of about 2,000 villages and 12,884 hamlets (out of the 39,000 hamlets listed in their topographical records). Most of the name changes occurred in the East and Southeast (in Mardin, 91 percent of the place names were changed), but significant changes also took place in some provinces in the Black Sea Coastal Region (in Trabzon, 72 percent of the place names were changed). The process of village renaming was put on the agenda again by the military junta after the 1980 coup and more changes enacted by the reconvened Commission in 1984, following the election victory of Özal’s Motherland Party, even though up to 90 percent of place names in some of the southeastern provinces had already been changed ( for an extensive discussion, see Öktem 2003; 2005: 185–221). In some cases, the ‘new’ names expressed a sense of religious, political, or ethnic identity. The name İslamköy (Diyarbakır-Kulp) evokes a religious identity and was attached to a settlement in an area formerly inhabited by Armenians; the name of the model village Cumhurriyet (administratively dependent on the central district of Diyarbakır) expresses republicanism; and Türkmen (Diyarbakır-Çüngü ) arouses the idea of an ethnic identity. However, such names seem to be exceptions. In most of the cases, the names do not have any pronounced religious, political, or ethnic significance, but refer to a general category from nature. In the Çınar district in Diyarbakır the villages were given ‘new’ names such as Gümü ta (Silver Stone), Ağaçsever (Tree Lover), Akçomak (White Cudgel), Ovabağ (Plains Orchard). These were not translations from the ‘old’ names. The Kurdish village Ku tiyan (Killed People) was renamed Soğansuyu (Onion Water); Kanipanık (Flat Spring) was renamed Yarımka (Half Eyebrow); and Bımbareki (Holy) was renamed Halkapinar (People’s Spring). In official publications and maps (the state’s representation of space) the ‘old’ names have been expunged and only the ‘new’ names are shown, but in daily life (the mental mapping or representations of space) the name changes did not have much impact. The authorities even put up signs with the new village names at the entrance to each village, to remind the inhabitants and visitors of the ‘new’ name, but
introduction
33
locals continued to call their villages and the surrounding villages by the ‘old’ names. Most of these signs have rusted away over time now. During my research, I had no difficulty in finding villages if I happened to know the ‘old’ names, yet I often lost my way when traveling to a village which I only knew by its ‘new’ name (which I had read about in official documents). This was not because of a lack of map readings skills or knowledge of the area where I was traveling. True, available maps were not always very detailed and tended to show only the larger villages, not the many small villages and hamlets, but on several occasions I had company of someone familiar with the area and we still had great trouble reaching our destination. The local people we met on the way could not help us if all we knew was the ‘new’ names—except for their own village and maybe a neighboring village, villagers simply do not know the ‘new’ names. Indeed, government institutions with local functions to execute, such as the Office of the Kaymakam (administrator of a district), do have access to village/hamlet information files which include both the ‘new’ and the ‘old’ names. A Note on Numbers Statements about numbers have to be considered with care for a number of reasons. Sometimes numbers are downplayed, while on others they may be inflated. Displacement and return figures are discussed in the coming chapters. Another case in point is the ownership of land. On several occasions landowners confided in me that they never gave a correct statement to government officials about the quantity of land they owned, and neither do I know if they ever gave a correct statement to me. The owners let me know that they were aware of the real reasons for official enquiries, and their possible effects (e.g. taxes and possible expropriation). There are, of course, other ways to seek to acquire such information, most obviously by consulting the Land Register. However, not all land is registered to the legally entitled person and in some cases land is not registered at all. Additionally, in spite of legal regulations stipulating that men and women are equally entitled to inherit land, in reality women do so only in rare cases (many women sign documents under pressure in which they relinquish their rights, and even if land is registered in a woman’s name, this is generally just a formality). The Ministry of Agriculture has operated its own database on land ownership since 2001, when, as part of an agricultural support policy, it began providing a per dönüm subsidy to farmers. Yet evidence
34
chapter one
suggests that these numbers are often, if not routinely, inflated, so as to obtain greater subsidies.16 It proved to be informative to ask the obvious, which turned out to be less obvious than it seemed. On one occasion I talked to a landowner in Diyarbakır province about the population of the place in which he lived, a small, rural settlement. We were standing on the roof of his house, with a view of the hamlet and beyond, when I asked him how many houses or households the hamlet comprised. The landowner’s reply—three houses—astonished me, as it was perfectly evident that there were more than three houses in the hamlet. From where I stood I could see at least six houses, and from the satellite dishes on the roofs it was clear all were inhabited too. As we entered into the subject further, I discovered that the landowner had only counted the houses inhabited by people with land entitlements. For the landowner, only those who owned land were looked upon as villagers. The houses inhabited by farm laborers and their families were not counted, even though some of them had lived in the hamlet for years and their children born in the village: because farm laborers reside in the hamlet as workers attached to a landowner and live in shelters provided by the landowner, landowners do not consider the laborers to be ‘inhabitants’.17 Leaving the issue of numbers in relation to land ownership, figures provided by government sources must be assessed critically too. For example, and critically for our subject, official statistics suggest that villagers have headed back to their homes in large numbers. These figures are of doubtful provenance, however: they tend to be inconsistent and almost never list the actual rural settlements to which villagers supposedly returned (an issue discussed in Chapter 6). Furthermore, it is common knowledge that the official population figures for most cities, towns, and villages are too high—I even had this confirmed by an employee of the State Institute of Statistics, DİE (Devlet İstatistik Enstitüsü). This
16 A director of agricultural affairs (tarım müdürü) in one small town narrated the story of how he had once received the local head of the village guard militia in his office. The man wanted the agricultural director to record his entitlement to subsidy at 500 dönüm, which would translate to a subsidy of about 5,000 euro, although his land registration certificate showed only 50 dönüm. To give weight to his claim, the local militia leader arrived with the support of the town’s army captain ( yüzba ı). 17 To illustrate the status of such workers, not only do farm laborers have to vacate their shelters if the labor contract is ended, but if they die it is not unusual for the wife and children to be forced to leave the village (generally to move to male relatives, probably the widow’s father or a brother).
introduction
35
may be due to exaggeration by municipalities or muhtars, attempting to get more resources from the central government (such resources being partly determined by population counts). Also, individuals return to the smaller settlements of their families from the larger settlements where they live and work during census taking, as a civic duty to their homelands for the same purposes of enabling a greater claim for state resources (muhtars will, for example, network and even organize buses for people to travel in, ‘back’ from the big cities to ‘their’ towns and villages in order to facilitate this).18 The temporary migration back to homelands at census times obviously skews official figures in ‘favor’ of towns and villages (and is one reason why the real population of İstanbul is typically estimated to be as much as 30–35% higher than the official figure). Another, contrary, factor regarding population figures is the seasonal difference there may be in the number of people resident in a small rural settlement. It is common practice for people throughout Anatolia to visit or return to villages and hamlets during the summer months, especially if these are not particularly hospitable places in the winter due to infrastructure shortcomings or elevation/cold weather, etc. This phenomenon tends to skew official figures ‘against’ villages, by ignoring the vast increases on permanent populations in the summer. Another numerical problem pertinent to the subject of this work is the difficulty of attaining a total figure for rural settlements in Turkey as a whole, including the Southeast. There are village statistics, but numbers mentioned in different sources vary widely and there appears to be a lack of clarity when distinguishing between villages and hamlets.19 (Upon inquiry, the State Institute of Statistics responded that it 18 Regarding the sense of ‘civic duty’, the traditional ties to homelands remain strong even for city dwellers, whose metropolitan urbanization more often than not has a history of less than two generations (i.e. either they or their parents moved from villages and towns to the city—when asked where they are from, such people will usually give the name of their homeland province or town, even though they might never have actually lived there). In the context of nationalist struggle and national identity, this cultural phenomenon of ‘expatriate’ attachment to the homeland obviously takes on an extra dimension for Kurds in Turkey (especially for those living in cities outside of the Kurdistan region), and a further dimension still if the settlements of their homelands no longer exist. Certainly, this must contribute to a sense of alienation, and foster the conditions for some young Kurds to become politically radicalized. 19 In a 1983 report by the State Planning Organization, DPT (Devlet Planlama Te kilatı), the number of villages is put at 36,155 and the number of both villages and hamlets at 93,553 (Güzelsu 1983a). However, in a 1984 DPT report, the number of villages is said to be 36,155 and the number of villages and hamlets 88,553 (Dülger 1984). Again
36
chapter one
does not count hamlets, and I was referred to the General Directorate of Village Services, KHGM (Köy Hizmetleri Genel Müdürlüğü), but without success.) It is also difficult to compare numbers used in secondary sources. For example, Gerard Chaliand (Chaliand 1993) states that the names of 12,861 of the 34,957 villages had been changed by 1980, and that some 80 percent of these changes had occurred in southeast Turkey. This implies a change of name for more than 10,000 villages in southeast Turkey, but according to Doğanay (Doğanay 1993) there were only 9,014 villages in the entire area in 1984. Controversies about numbers most likely results from different systems of counting, and pose serious problems in using and comparing statistics, from different sources of course, but even, in the case of the State Planning Organization, from within a single source. A Note on Data Collection The State carries out research but tends not like to be an object of research itself. This is definitely true in the case of the Republic of Turkey, and posed serious constraints on the collection of data. Information on resettlement projects was particularly difficult to obtain. Both the Southeast Anatolia Project, and the General Directorate for Village Services, institutions involved in settlement and resettlement projects, did not provide easy access, if any at all, to information on studies conducted or projects implemented. Not only were official reports difficult to obtain, but also the collection of field data was hindered by limited access to villages. The following anecdote illustrates the types of problems encountered in this respect. On April 19, 2003, I set out on a journey to İslamköy but was stopped at a military checkpoint and, after the soldiers on duty had consulted their commanding officer by telephone, sent back to Diyarbakır. The soldiers told me I was not allowed to go to the Diyarbakır districts of Lice, Hani and Kulp—İslamköy is located in the latter district—and stressed that I had to ask permission from the Governor. On my return I twice submitted a formal request at the Governor’s office for permission
another DPT report fixes the number of villages at 35,023 and villages and hamlets on 79,342 (Doğanay 1993). In the year 2000, a DPT commission counted 73,000 rural settlements, including 35,427 villages (DPT 2000).
introduction
37
to visit İslamköy, but my requests were turned down on the grounds that İslamköy and the districts I had to travel through—notably Lice and Kulp—were restricted areas and therefore forbidden to foreigners. The Governor’s assistant advised me to pay a virtual visit to İslamköy through the websites of the Governor and the Southeast Anatolia Project. This was, in fact, the second time that I had been advised to gather my data through the Internet. Some months before, the World Bank had also referred me to its website for information on settlement projects in Turkey that it was considering investing in. Perhaps the virtual world is better ordered to meet the foreign visitor’s gaze than reality. On the Governor’s and GAP websites, the İslamköy project was presented as a successful example of resettlement and rural development, which intrigued me even more. Hoping for the best, I made further attempts in 2003 and 2004 to see for myself, but was turned back at the same checkpoint on both occasions. Eventually, just days before the end of my last fieldwork period in June 2004, I did manage to visit the village and meet the villagers in their own setting rather than in a coffeehouse in Diyarbakır city. Foreign researchers have been hampered by the inaccessibility of the region and its villages for a long time. Martial Law and State of Emergency regulations have ruled the Diyarbakır area almost continuously since 1928. It was declared a military zone forbidden to foreigners until 1965, and has been administered since 1978 under a Rule of Exception. In addition to that, the area was ravaged by a war between Turkish armed forces and the Kurdish guerrilla, the PKK, through the 1980s and 1990s, during which thousands of villages were razed down to the ground and large parts of the countryside turned into military zones and no-go areas. Since the State of Emergency was lifted in November 2002, field access has continued to remain very limited to foreigner investigators, to wit, my anecdote above. Just as problematically, decades of war have produced a lack of trust among and between people, sometimes even between families and neighbors, so how much more suspicion will settle upon a stranger seeking to probe sensitive issues as village evacuation and return. This was compounded for me in that my language of inquiry was Turkish, as I did not master the Kurdish language. Enquiries in Turkish would occasionally raise doubts as to who I was. On some occasions I was held to be a subay (an officer in the army). Although this mistrust could in most cases be overcome by getting acquainted, fear of the authorities did
38
chapter one
not always abate. Some people especially who had recently returned to their villages were afraid to talk, concerned in case this might upset the delicate balance between themselves and the authorities, and preferred to avoid contact with me altogether—which, obviously, I respected. The empirical data in this study suffers from the limitations here described. Conversely, however, these restrictions and limitations also constitute evidence. The absence of information or refusal to speak also conveys meanings of a kind which one should endeavor to interpret. Structure This study covers three basic issues. The first is the rationale of resettlement policies of Turkish State institutions in the context of the war during the 1990s, when the predominantly Kurdish east and southeastern (Kurdistan) part of Turkey was the scene of an extended and organized violent conflict between Turkish armed forces and the PKK. State forces evacuated and destroyed rural settlements on a large scale, resulting in the resettlement of a major part of the rural population in urban areas, and leading to the development of several plans for a re-design of settlement structures in the countryside. This work discusses how these resettlement policies were supposed to re-organize social life and with what aims; it also reveals contradictions between different settlement practices and how these contradictions are related to different representations of space.20 The second area of interest is the relationship of resettlement policies from the 1990s with settlement policies and plans from earlier times, policies and plans going back to the 1960s and 1970s, and even the 1930s. This gives a historical perspective to the analysis of resettlement. The third issue taken up is the relationship between particular ideas about what modernization stands for, rural modernization in particular, and its implications for (re)settlement policies and identity politics. Analytically, resettlement practices are related to processes of nation- and state-building. 20 It is acknowledged that the PKK was also responsible for some evacuation of rural settlements in the period 1985–1990, but only, in the main, those which had been recruited in the paramilitary system; i.e. PKK guerilla fighters attacked and forced the evacuation of some rural settlements whose inhabitants had joined the Village Guard system and become militiamen for the state (Yıldız & Hughes 2003: 18). The evacuations of rural settlements by the PKK are beyond the scope of this study (although it is touched upon in Chapter 4).
introduction
39
The body of the book is composed of five chapters. To some extent these chapters are arranged counter-chronologically. Chapters 2 and 3 discuss the institutional rationale of resettlement practices in Turkey and their execution (or non-execution) during and since the 1990s. Chapters 4 and 5 move from an analysis of the inherited spatial/settlement patterns and current administrative set-up to a genealogy of concepts used in resettlement policies, a genealogy that follows the break up of the Ottoman Empire and establishment and entrenchment years of the Republic through to the 1960s and 1970s, during which the foundation concepts of contemporary polices were laid.21 Chapters 4 and 5 thus add a historical dimension to the social analysis of resettlement; they also link resettlement to the more inclusive concept of settlement. Chapter 6 returns to the end of the 20th and first years of the 21st centuries, with information and analysis from case studies, and discussion of developments ‘on the ground’. The counter-chronological format is prompted by the immediate concern of this study: an understanding and social history of the resettlement concepts and practices of state institutions in Turkey in the context of the war with the PKK. The historical perspective afforded by the genealogical approach of Chapters 4 and 5, reveals two different kinds of modes of thinking on the spatial dimension of modernization—with, in addition to the industrial city, a new kind of settlement unit in the countryside, at the same time both urban and rural, a spatial conception that had a profound impact on settlement policies in Turkey. The analysis of resettlement will start with a discussion of the background of resettlement in southeast Turkey in the 1990s, which is a PKK led insurgency and State-led counter-insurgency. Chapter 2 considers the PKK guerrilla strategy in relation to the concepts of spatial and social environment. It is argued that the PKK initially exchanged a student-urban environment for a popular-urban environment in the Southeast. After the coup of 1980 and the decision to organize a
21 It is difficult to use the concept of genealogy without referring to Foucault, who introduced the idea of a ‘genealogy of knowledge’. Foucault’s genealogy of knowledge consists of two bodies of knowledge: first, dissenting opinions and theories that did not become the established or widely recognized, and second, local beliefs and understandings. Foucault says: ‘Let us give the term “genealogy” to the union of erudite knowledge and local memories which allows us to establish a historical knowledge of struggles and to make use of this knowledge tactically today’ (Shawver 1999). In this study, although genealogy refers simply to ancestry, it does also reveal a struggle between different state and elite groups about how to develop territory and organize a nation’s space.
40
chapter one
guerrilla war, the PKK then chose a villagers-rural environment for the purpose of building up forces under the protection of a guerrilla army. The chapter will explain how the army strategy of resettlement was designed to contribute to environment contraction and deprivation by depopulating those rural areas where the PKK had established itself. The main proposition of this chapter is that the resettlement of villagers was not collateral damage, but one of the main constituents of the military’s counter-insurgency program. The issue of return entered the political agenda almost simultaneously with the evacuation of villages. Although Turkish politicians were inclined to consider return conditional on approval by the military, the implicit assumption was that the evacuation of the countryside was an anomaly, and return inescapable. In Chapter 3, the most important plans of civil authorities for the development of a framework in which return was supposed to occur are discussed, namely the village-town and center-village approaches. Consideration of these plans reveals differences in institutional thinking between political and military authorities. The main proposition of this chapter is that the return-to-village plans are in sharp contradiction to the objectives of the spatial policy of the military and the governors, and institutional disagreement between various state-institutions are the most important barrier against the implementation of any real plans at all in the region, including those of return, leading partly to a generalized ossification and territorial definition by proxy, a victory for the military strategy of indefinite duration as a fait accompli, but partly also in spontaneous and novel patterns of (re)settlement. The origins of the spatial concepts used for resettlement and settlement practices (center-village, village-town, and village abolition) are discussed in Chapter 4. The genealogy of concepts reveals a quest towards a new type of settlement in the countryside, allegedly responding to the demands of modernity, i.e. by allowing both vertical integration, a means for down-scaling civil administration and cultural identity, and also economic integration, through a refined division of labor. The quest for a new settlement type (by means of the abolition of small rural settlements and their clustering in village-towns or center-villages) is discussed against the background of (what the state and nation-builders considered to be) two major problems: the irregular distribution and large number of small settlement units in the countryside, and the need to downscale administration. Concepts such as the center-village
introduction
41
and village-town were intended as tools for the reorganization of rural space, integrating the small rural settlements into the national administration, economy and culture. A central proposition of this chapter is that settlement policies are influenced by two rather different, but not incompatible, forces: the desire to expand administration to rural areas and an alternative view of urbanization (allegedly producing Turkishness), which detaches urbanization from the prevailing villageto-city migration patterns and instead aims at the urbanization of the countryside. This was the logic in the planning of the new village-town and center-village settlement types. In Chapter 5, the desire to develop a new settlement type is related to ideas about nation-building emerging in the 1930s. More specifically, this chapter will explain how, in the establishment and entrenchment years of the Republic, the quest for nation-building tools produced the idea of a settlement type beyond the modern rural-urban divide. This new settlement type, referred to as rurban, was not supposed to contribute to vertical integration (into urban centers, as was the case with the new settlement types developed in the 1960s and 1970s), but rather to wipe out the rural-urban divide altogether. A main proposition of this chapter is that resettlement has been reinvented as a means for the production of a population with the right characteristics. It has, so to say, become an instrument of a Gellnerian type of modernity, employed for the creation of a culture-based polity. Although obstacles in the way of return have been documented at length, return itself has not been the subject of previous study. Chapter 6 discusses developments on the ground, the actual practices of resettlement. The data here is mainly collected by means of case studies. The chapter gives an indication of the magnitude of return, the role of the authorities, and the difficulties returnees have to cope with. Additionally, an analysis of the cases will show ambiguity in the idea of ‘return’, not only because new (and unplanned) settlement patterns and structures are emerging, but also because some ‘returnees’ are not, in fact, former inhabitants of the settlements, but new settlers. The main proposition of this chapter is that the inhabitation of evacuated areas is not leading to a recovery of old settlement structures and patterns. In the areas affected by village evacuation and destruction, fragmentation and hamletization is occurring at the structural level, and novel forms of extended settlement styles arising at the level of settlement patterns. Interestingly, these new structures and patterns are in contradiction to the objectives
42
chapter one
of both the resettlement policy of the military and the return plans of civil institutions, as discussed in Chapters 2 and 3, and the very aims of settlement policies as discussed in Chapters 4 and 5. In the concluding chapter, the analysis of resettlement is related to the more general questions of war and modernity. The discussion focuses on policy issues in Turkey, in particular Turkey’s Kurdish issue, which is at the foundation of recent resettlement practices.
CHAPTER TWO
SOLDIERS AND SETTLEMENTS Turkish Counter-Insurgency and the Strategy of Environment Deprivation Where there is sea there are pirates. In this province [Hakkari] are 674 villages and hamlets. These settlements form the spider’s web in which the PKK feeds itself. (. . .) [W]hy don’t we concentrate all [villagers] in two or three main settlements. General Osman Pamukoğlu1
Introduction The notion that a small group of politically highly motivated but relatively poorly equipped guerrilla fighters could defeat the second largest armed forces in NATO in a prolonged war seems incredible. But various attempts by Turkish politico-military organizations to prosecute a counter-insurgency guerrilla campaign against the separatist PKK during the second half of the 1980s had failed dramatically, resulting in the deaths of their leaders and main cadre, and by 1990 the ‘liberation of Kurdistan’ by the PKK had become not at all unthinkable. The Turkish Armed Forces were rapidly losing control of an undeclared war. They only managed to regain control several years later, after a reorganization of the army and the implementation of a radical (counter-insurgency) war doctrine. The ‘new’ war was characterized by a harsh resettlement practice, with forcible, wholesale evictions accompanied by summary executions, the slaughter of livestock and the burning of villages. In this chapter, we discuss the military logic for this resettlement. The objective, it is argued, was not merely retaliation, and nor was it a negative side-effect of the war or collateral damage. It was not even a temporary clearance of the countryside in order to break the guerrilla. Rather, the resettlement program constituted a concerted attempt
1
Pamukoğlu 2003: 59.
44
chapter two
by the Turkish military to bring about a conclusive transformation of the regional settlement structure. Its rationale was spatial contraction leading to environment deprivation. The small rural settlements of the Southeast were regarded as the ‘negative environment’ that fostered the PKK guerilla, and the immediate aim of the Turkish Armed Forces was to solve this problem by changing the spatial ground of the war. The destruction of rural society was simply considered part of the solution. A top secret letter written in 1993 by President Turgut Özal reveals some of the basic characteristics of the counter-insurgency strategy that began to emerge in Turkey between 1991 and 1993 and was fully implemented in the years after. A key element in this letter was his urging of the need to reorganize the space of war, a need that followed from the spatial strategy of the PKK Commencing with excerpts and consideration of Özal’s letter, this chapter goes on to look at the origins and strategy of the PKK. The discussion focuses on the PKK’s spatial strategy, which was marked by the decision of the PKK to detach itself from the urban student environments in western Turkey and develop an insurgent strategy that attached major importance to the building up of military forces in rural areas in the Kurdistan region in Turkey. It is against this background of insurgent strategy that the changing counter-insurgency tactics of the Turkish Armed Forces are discussed. We consider how the state military strategy up until 1991 created space for maneuver for the PKK, enabling the organization to establish itself firmly in the countryside; and how the radical change in strategy was developed to combat the PKK guerilla, aimed at environment destruction and based upon a two-pronged approach of i) the penetration of ‘PKK spaces’, with the use of special forces and peasants villagers seconded as paramilitary ‘village guards’, and ii) spatial contraction by means of resettlement of the rural civilian population—which ultimately resulted in field domination by the Turkish Armed Forces. The Özal Proposal Turgut Özal had been Prime Minister of the Republic of Turkey between 1983 and 1989, the period in which the PKK had developed into a force capable not just of causing the Turkish state problems, but of actually winning the struggle for independence from Ankara and achieving the establishment of a Kurdistan. In 1989 he ascended to
soldiers and settlements
45
the presidency, as the PKK rose to the height of its power, effectively controlling great swathes of ‘Turkish’ territory. Özal, as much as anyone, was aware of how grave the situation had become. Shortly before his sudden death on the 17th of April 1993, the president wrote a topsecret letter to the then prime minister, Süleyman Demirel, proposing a solution to the PKK insurgency. This letter, with the proposals listed, symbolizes the birth of a new war in Turkey. The following parts of the letter are taken from a translation published in 2002 by the Kurdish Human Rights Project (KHRP 2002: 118–22): ‘. . . In the Southeast, we are faced with perhaps the most significant problem in the republic’s history. The ‘Kurdish Question’ in south-eastern Turkey, what with its political, social and economic aspects, and with bloody acts of terrorism, poses an ever-growing danger. The beginnings of the problem date back to the final years of the Ottoman rule. In the 15 years that ensued after the declaration of the republic, the state had to put down a number of rebellions [ by Kurdish secessionists]. Blood was shed when necessary, and a certain portion of the local population was forced to migrate to the west of the country. ‘With the annulment of a policy of forced migration following the introduction of democracy in 1950, some of those forced to settle in the west returned. Yet starting from the 1960’s, the local population again began shifting towards the west. Despite the lack of definitive official figures, 60 percent of those called Kurds probably live in sectors of the country west of Ankara. Because the migrations were not planned ones, in certain provinces in the West—such as Adana, Mersin, Izmir, Antalya, and even Istanbul—our Kurdish citizens—live in close proximity in certain districts. ‘The problem we face is way beyond the simple dimensions of terrorism. Therefore, it is imperative to consider short-, medium-, and long-term solutions and to adopt two separate approaches for dealing with the local population and the terrorists. ‘Despite the availability of information on the causes of the problem, no in-depth analyses have as yet been made. In order to add to the efficiency of the policies we have been pursuing, our struggle against terrorism must be backed by comprehensive analyses by scientists, both foreign and Turkish. Research groups should immediately be set up with a view to conducting investigations on socioeconomic and psychological aspects of the issue. Public opinion polls should be conducted to improve understanding of the problem. Research groups should comprise scientists, state officials, statisticians, soldiers, and other relevant experts. ‘It must be borne in mind that owing to military measures being taken to wipe out terrorist activity, the locals in the Southeast have been subjected to harsh treatment and feel, as a result, estranged. If there have been mistakes made in tackling terrorism, they should be frankly discussed and realistic solutions must be sought.
46
chapter two ‘A complete overhaul of the training system of security forces is necessary. This should be accompanied by the modernization of their equipment and of the methods they employ to fight against terrorists. They need re-education in “public relations”. ‘Starting with the most troubled zones, villages and hamlets in the mountains of the region should be gradually evacuated. With this group of PKK (outlawed Kurdistan Workers’ Party) supporters, in number no more than 150,000 to 200,000, being resettled in the Western parts of the country according to a careful plan, logistic support for the PKK will have been cut off and their standard of living will have improved. This group should be given employment priorities. ‘With the evacuation of mountain settlements, the terrorist organization (PKK) will have been isolated. Security forces should immediately move in and establish complete control in such areas. To prevent the locals’ return to the region, the building of a large number of dams in appropriate places is an alternative. ‘On all highways in the region, 24-hour patrol duty by special teams is a must. Helicopters in day-time, and night-vision armoured personnel carriers at night, must be on patrol duty. A complete overhaul of the security network in the region is urgent. Security personnel must be transformed from a defensive force to one that is offensive. ‘The purchase of 20 Cobra and 20 to 30 Sikorsky helicopters for the security forces deployed in the area will help create a mobile force that can handle incidents that might occur simultaneously. The restructuring of state intelligence organizations active in the Southeast is an urgent priority, to make up for lack of sufficient information on the (PKK) plans. Coordination must immediately be effected between the National Intelligence Organization (MIT), the gendarme’s intelligence command, the armed forces, and police. ‘A 40,000 to 50,000-strong special force, comprised of full professional units, with at least one year of special training behind them, should be set up to fight against the PKK. They should be paid satisfactory salaries. Unit commanders in this force should be given leeway to take initiatives on any issue when conditions necessitate it. The special force must not be a force on the defensive. It must be a force that tracks terrorists down and attacks them. Naturally, they should maintain contact with other units deployed in the area and cooperate with them. Ordinary units of the standing army must only be used for routine military duties such as security checks and control. ‘Border trade, an important source of income for the local population, must be free. The opening of new border posts with Syria, and the reopening of those that have been closed, are necessary. An improvement in border trade will mean new opportunities for the locals and make life easier for at least some. ‘In order to cut off logistic support for the PKK, the local people should be won over to the side of the state. The people settled in faraway mountain villages and hamlets should be encouraged to move into bigger settlement areas.
soldiers and settlements
47
‘Given a tendency for the locals to migrate to the west of the country, it would appear that only 2 to 3 million people will inhabit the region in the future. If this migration is not regulated, only the relatively well-off portion of the population will have moved and the poor will have been left behind. Thus the area will turn into a breeding ground for further anarchy. To prevent this, the migration must be regulated by the state. A planned, balanced migration, including members from all segments of society, to predetermined settlements in the West is essential. ‘In addition to committing terrorist acts, [the PKK] is spreading widespread, effective propaganda with the purpose of intimidating and ultimately brainwashing the local people to win them over to its side. Counterpropaganda to strengthen local support for the state, to boost morale, and correct disinformation is of crucial importance. ‘Therefore, it is imperative that special efforts be spend to inform both the public and the international community of the true nature of developments. In order to do this, the setting up of a special team of experts to create a favourable c1imate of public opinion is necessary. Thus the scope of our activity in releasing press statements, leaking news, and, if need be, spreading “disinformation” will increase. ‘It is of the utmost significance that the statements made to the press regarding the security forces’ struggle against terrorists be regulated with the greatest possible care. Press reports, both written and visual, which could be exploited by [the PKK] to highlight itself as either a ‘heroic or an innocent’ organization, must be avoided. Medium- and long-term suggestions ‘Such cities as Adiyaman, Diyarbakır, Urfa, Mardin, Siirt, Elazığ, Malatya, Erzincan, Erzurum, Kars, Ardahan, and Iğdır must be turned into centres of attraction for the local population currently settled in the countryside. This should be done through special incentives for investors. Thus, the evacuation of the countryside will have been facilitated. ‘Incentives must be provided for the private sector to invest in the region. Corporate tax should be lifted for a long period. Income tax levied on the locals must be decreased and the electricity supply must be cheapened. ‘This problem should be debated freely, in an unbiased manner, in the prejudice-free atmosphere. Through debate, the rights and wrongs will come to light, thus leading us closer to the truth. To bar discussion, to cover up the truth, will not alleviate the problem. On the contrary, it will lead to further chaos because of the adoption of a mistaken approach. Conclusion ‘If mistakes are not committed, and inconsistent, unnecessarily hurried action is not taken, the fire in the Southeast will die out in five to ten years, with the weakening of nationalist sentiment and the decrease of foreign involvement (Because the fire has undoubtedly been started and fuelled
48
chapter two by foreign powers which desire to prevent Turkey from using its historic opportunity to accomplish its aim of becoming a powerful nation.) ‘State officials must not project an image that shows Turkey as a country afraid of, and intimidated by, terrorism. It will be of great help to show to the world outside Turkey that the state is capable of tackling such an issue and that it is not at a1l wary of, or worried about, terrorism. ‘Therefore, it is the responsibility of all state officials, whatever their rank may be, of politicians, and of the press, to differentiate between terrorists and the local population and to treat them accordingly, in order to maintain our unitary state apparatus and to wipe out terrorism by taking the above-mentioned measures to prevent its incidence. . . .’
President Turgut Özal’s proposal was marked by some striking features. First, the problem was defined as a spatial problem. A direct relationship was made between an effective combat of the PKK and the evacuation of villages and hamlets in the troubled areas and the resettlement of their populations. The clearing of the countryside was supposed to contribute to an isolation of the PKK, and the development of large infrastructure projects, such as dams, to prevent the return of deportees and a recovery of an environment hospitable to insurgency. Second, a relationship was drawn between an effective war against the PKK and knowledge about the people living in the insurgent areas. Knowledge about people was supposed to contribute to the practicing of control over them. The transformation of the insurgent region and its population into knowable objects became a growing interest of the authorities after 1989 (Özok 2004). The task of turning the region and its population into object of ‘knowledge’ was mainly allocated to the so-called Southeast Anatolia Project, GAP (Güneydoğu Anadolu Projesi).2
2 Symbolically, water has always been seen as the spring of life, and the development of water-resources by means of the Southeast Anatolia Project is considered vital for the advancement of Turkey. The Southeast Anatolia Project GAP is a focus for industrialization in Turkey and of modernization of the Southeast. Water resource investigations in the region were initiated with the establishment of hydrometric stations on the Euphrates and Tigris Rivers in 1936 and 1947 respectively. Topographical and hydrological maps of the region were made, reconnaissance studies completed in 1958, and studies to assess the energy potential completed in 1963 (Lorenz & Erickson 1999). Construction of the first major dam in the area (the Keban Dam) began in 1965 and was completed in 1975. In total, the project envisaged the construction of 22 dams, 19 power plants and hundreds of kilometres of irrigation canals. In quantitative spatial terms, the GAP region covers a surface of more than 75,000 km2, corresponding to almost 10 percent of the total surface of Turkey, yet the 1.7 million hectares of arable land served by the project is about 20 percent of the total irrigable land in the country. The population in the region is 6.6 million people, which is slightly less than 10 percent of the national total. Although GAP did originally start purely as an
soldiers and settlements
49
The GAP project can be properly described as a regional development administration aimed at the socio-technological transformation of eight provinces in the southeast of Turkey, bordering Syria and Iraq.3 Between 1992 and 1994 several studies were done that aimed to develop a better understanding of the make-up of the region, including its cultural composition. Studies were carried out on trends of social change in the GAP region (Sencer 1993), on employment and resettlement (Birsen 1994), population movements (Ak it 1994) and the (changing) status of women in the region (Saltık 1994). In these studies, a description of the population was related to ‘structural trends’, which were thought to mark the process of long term development of the region (and would alter its socio-cultural make-up). The studies also called for a more liberal cultural politics in the region (Ak it 1994),4 and better governance (Birsen 1994).5 energy and irrigation project intended to realize the potentially rich water and land resources of the region, it had been expanded to enter the fields of rural and social development by 1989. 3 The GAP area does only comprise the southern part of the Kurdistan region in Turkey, covering eight provinces (Adıyaman, Batman, Diyarbakır, Gaziantep, Siirt, anlıurfa, Mardin, and ırnak). The northern part of that region falls under the authority of the Eastern Anatolia Project, DAP (Doğu Anadolu Projesi), covering 16 provinces (Ağrı, Bingöl, Bitlis, Elazığ, Erzincan, Erzurum, Gümü hane, Hakkari, Kars, Malatya, Mu , Tunceli, Van, Ardahan, Bayburt and Iğdır) never gained the momentum GAP has. 4 Without explicitly mentioning Kurds or the Kurdish language, Aksit writes in the executive summary: ‘The existence of different sociocultural identities has been confirmed by the present research as well. Various policy proposals can be developed from this sociocultural reality. Researchers working on sociocultural identities have shown that such identities are formed in the context of language. A plurality of such sociocultural identities exist among the household heads in villages and cities of the region and among the migrants to the metropolitan areas in the West. Existence of plurality of such identities have been treated as sources of cultural richness in many of the modern countries. (. . .) It has to be pointed out that plurality in sociocultural identities and languages have also consequences for extension and community participation programs. These should be taken into consideration while planning and implementing such programs.’ 5 Also in the executive summary, Birsen Gökçe writes: ‘The existing practice in the region is shaped in a manner to exclude local governments and to limit solution finding to the central authority. Being already weak and at its infancy in the region, local governments have been almost left aside in GAP implementation. As a result, participation and commitment of the local pepole could not be realized.’ And: ‘The process of expropriation and resettiement has once more laid bare the traditional structure of administrative authority—individual citizen relationship in Turkey. The tradition which has strong resistance to sharing any part of central authority with others has also persisted in this process which excluded any form of participation. The people of the region who are already lagging behind country averages in terms of educational status and means are almost helpless and alone against the cold and authoritarian
50
chapter two
Thirdly, a relationship was drawn between a reorganization and modernization of the Turkish Armed Forces and successful counterinsurgency. The Turkish Armed Forces had been applying methods of conventional warfare, based on an outmoded doctrine. The cold war had come to an end, but even after the fall of the Berlin Wall, the military strategy of the Turkish army was continued to be based on the doctrine of static territorial defense by a standing army against a coherent mega-threat (an invading Soviet Army, centrally organized and clearly identifiable). The war with the PKK did not fit the doctrinal remnants of the cold war, however, and the army was rapidly losing control. The PKK guerrilla did not fight classic pitched battles. Instead, it was characterized by poly-centricity (the ‘people’s army’ had no center, but was composed of innumerable relatively independent units, clustering with and separating from other units whenever deemed necessarily); multi-directionality (the guerrilla army was difficult to localize and there was no front, its militants were permanently on the move and giving the impression of being able to attack at any time and any place); and multi-dimensionality (not all party militants were guerrilla, many were active in the political ‘front’ and within civil society organizations such as trade unions). Özal proposed a reorganization of the Army so as to be able to wage an asymmetric war—which constituted a revolution in the Turkish military and military philosophy of the time. In the years that followed the letter, most of the measures proposed by Turgut Özal were enacted, such as the idea to deprive the guerrillas from their social environment (by means of the resettlement of its supporters), and the penetration of the physical environment (rural and mountainous areas) by special units. Özal’s proposal to resettle people from the most troubled zones ‘according to a careful plan’ seems not to have been implemented. At least, the resettlement of the civil population of these areas was not carried out in the form of a scheme, defined as a systematic plan of action arranging for the physical removal of the people and their settlement in new locations, with compensation of losses, development of new means of support, and the allocation of resources for the execution of the actions and activities identified.
directives of the state which bothers for no further explanation or elaboration.” Gökçe concludes that the establishment of the institute of a “regional ombudsman” might be considered, together with ‘additional measures’ for the protection of citizens and their rights before the state administration (Gökçe 1994: 28–9).
soldiers and settlements
51
Nevertheless, although resettlement was not schemed, it was carefully planned in the form of tracks, as explained below. First, however, let us look at the enemy to whose success Özal was responding, the PKK. The PKK: Urban backgrounds and rural constitution Profile Unlike most Kurdish political parties, which adopted a rather conservative outlook and were organized around tribal leaders and structures, the PKK originated from the radical left in Turkey and drew its leaders, members and militants from the disenfranchised (the undisputed leader of the party, Abdullah Öcalan, was born in 1947 to a poor family in Omerli, a village in the southeastern province of Urfa, bordering Syria). The PKK’s fierce stance (the party did not hesitate to use violence, not only against the state but also against the powerful Kurdish tribal leaders), strong convictions, and disciplined but decentralized organization contributed to a steady rise and growing effectiveness of the party through the 1980s.6 After years of training, the guerrilla war for the political liberation of Kurdistan and a social revolution in Kurdish society was initiated with simultaneous raids on the gendarmerie stations and officers’ apartments in the Eruh and emdinli districts of Hakkari on the night of August 15th, 1984. When the PKK began its guerrilla campaign, the organization had no more than a couple of hundred armed fighters—within ten years this number had increased to 15,000–20,000.7 This figure alone indicates the success of the guerilla war against the Turkish State and its Army, and which forced Özal to propose his radical counter-strategy. The PKK was not, of course, employing a military innovation in the use of the guerrilla option; on the contrary, guerilla warfare has a long tradition, history is replete with examples of its successful (and 6 The PKK shifted perceptions of social reality in Turkish Kurdistan. Not only was the state denounced for its repressive politics and denial (inkar) of Kurdish identity, peasants also gained the courage to criticize feudal leaders, and women to denounce male domination. The local ideological effect of PKK propaganda was tremendous. 7 In his study of the PKK, Paul White (White 2000: 143) presents the figures of the US State Department, according to which the PKK had a guerrilla force in 1995 of about 15,000 also supported by a part-time militia of 75,000.
52
chapter two
unsuccessful) employment, and there were several recent and ongoing guerilla conflicts at the time to serve as models.8 At a global level guerrilla warfare had become a means for political change, and in fact innumerable Kurdish nationalist and Turkish socialist parties in Turkey thought that political change was contingent on guerrilla warfare. Nevertheless, until the PKK took up arms, actual military struggle was refrained from by most other Kurdish political groups, notably the Kurdistan Socialist Party, PSK (Partiya Socialîst Kurdistan).9 Earlier attempts that had been made by the Turkish left to engage in rural guerrilla warfare had failed dramatically, resulting in the death of their leaders and main cadre.10 8 Guerilla-type options were outlined by the military theoretician Sun Tzu two and half thousand years ago, as among the military principles enumerated in the classic ‘The Art of War’ (Sun Tzu 2005); the Roman Fabius achieved success against Hannibal by means of a guerilla military strategy; guerilla warfare was used by Arabs in defense against the Crusaders; and Napoleon Bonaparte’s regime in Spain was resisted by Spaniards using guerilla tactics—from whence the name: guerra (war) + -illa (diminutive, to indicate the asymmetric nature of civilians waging war against a state army). Guerrilla tactics (influenced by Sun Tzu’s work) had contributed to the victory of the Red Army in China under Mao Zedong, who in turn developed an influential rural guerilla (mobile) warfare theory (later adopted by the PKK—see following section) (Zedong 1961[1937]). Guerilla strategies had been successful during the twentieth century in nationalist struggles against the mightiest of (imperial) opponents, including Nazi Germany (where the Partisans of Yugoslavia liberated most of the country before the Nazi defeat); Britain (enabling the Irish to gain national independence and Jews to establish the state of Israel), and France (gaining independence for Algeria and instrumental in the development of Ho Chi Minh’s NLF during the First Indochina War) (Taber 1965). Contemporary, inspirational Marxist guerilla leaders, Fidel Castro and Che Guevara, emerged from the Cuban revolution, but it was the 1975 culmination of the NLF ‘Viet Cong’ struggle with the eventual victory of the North over US forces in South Vietnam (again following Maoist theory, and at a time when the PKK was in its formative stage) that constituted the most momentous modern example of the prime strategic strength of the guerilla: its success in resisting a foreign force (which Turkey certainly was in Kurdistan, from the Kurdish nationalist perspective). Guerilla campaigns (generally military means to realize a nationalist/separatist independence and/or Marxist-Leninist political agenda) were being waged in many parts of the world during the period of the seventies and eighties when the PKK was developing and coming to prominence, including for example, India/Bangladesh and Tamil Sri Lanka in Asia, Angola and Mozambique in Africa, Columbia and Nicaragua in the Americas. 9 Even the PSK, generally considered to relatively moderate, considered establishing a military wing in the early 1980s. Elsewhere, notably in Iraq and Iran, Kurdish political parties were also attracted to the politico-military approach. It was only in Syria that the Kurds refrained from military undertakings (although many Syrian Kurds joined the PKK and fought in its struggle against the Turkish army). 10 The leadership of the first guerrilla parties, The People’s Liberation Army of Turkey, THKO (Türkiye Halk Kurtulus Ordusu), established in 1970, and The People’s Liberation Party of Turkey, THKP (Türkiye Halk Kurtulu Partisi-Cephesi ), established
soldiers and settlements
53
The use of military means was prompted by the narrow political space. The Turkish state refused to accommodate Kurdish aspirations or enter into political discussions on the matter, refusing to even recognize the existence of Kurds and inventing the concept of ‘mountain Turk’ (dağlı Türk) instead (in the Republic, ‘citizenship’ was considered to be equivalent to cultural/ethnic ‘Turkishness’, and in practice Kurds were forced to qualify themselves thus, as cultural/ethnic Turks (Barkey and Fuller 1998: 10).11 A state of exception and the constant threat of violence established and maintained a particular level of State dominance over Kurds.12 However, the violence of power was answered by the violence of opposition, and the control of the Turkish State was seriously threatened when the PKK succeeded in gaining mass support and presenting itself as a viable alternative. The violence perpetrated by the PKK, it has been argued, was rational and instrumental in the sense that it tried to change the social and political status of the Kurds, either by independence or some form of autonomy, in circumstances where there was no alternative avenue of genuine political expression (Bozarslan 2004: 23). Today, the threat posed by the PKK to the political system in Turkey and the influence which the PKK has managed to achieve among Kurds in the Southeast is recognized as the most serious challenge yet to the Republic since it was established in 1923 (Özdag 2003: preface). Clearly, it would be wrong to characterize the PKK as a primarily military organization. Basically, the PKK is a political party that uses military means to achieve political ends. From the outset, the PKK did not proclaim the creation of a unified and independent Kurdistan to
in 1971, were killed through the passing of death sentences (Deniz Gezmis, Hüseyin Inan and Yusuf Aslan) and in a military operations in 1972 (Mahir Cayan, together with 10 other cadres of the party). 11 Cultural interpretations of Turkishness are dominant, but racial ideas have formed a strong undertone in Turkish politics (see, e.g., van Bruinessen 1997). 12 The southeast of Turkey, or the northern parts of Kurdistan, had been ruled under martial law and emergency regulations since 1927. Until 1952, the area, or more specifically Bitlis, Diyarbakır, Elazığ, Hakkari, Mardin, Siirt, Urfa, and Van, was administered by an Inspector General, an office established in 1927 to bring ‘order and discipline’ (Kocak 2003). In 1935, two further Inspector Generals were appointed to administer ‘Kurdish’ areas, one for the ‘Murat and Munzur’ region, covering Dersim (Tunceli), and the other for the northern part of the Southeast, covering Ağri, Çoruh, Erzincan, Erzurum, Gümüshane, Kars, and Trabzon (the two other Inspector Generals, there had been five, administered Trace in the Northwest and Antakya in the South). The Southeast was closed to foreigners until 1965, and the region subsequently ruled under martial law or state of emergency from 1980 (until 2002).
54
chapter two
be its sole goal, but also desired a radical transformation of the social and political organization of Kurdish society. In its 1978 manifesto, Kürdistan Devrimin Yolu (The Path of the Kurdistan Revolution), written by (or at least accredited to) Abdullah Öcalan, the PKK made itself known as a ‘national democratic’ and ‘revolutionary’ movement. A destruction of colonialism (not only Turkish colonialism, but also the colonialism of the other occupying state-forces in Kurdistan) and the construction of a democratic and united Kurdistan, based on MarxistLeninist principles, were to be effected from an alliance of workers, peasants and intellectuals.13 During the course of its existence, Abdulah Öcalan tried to develop an original understanding of socialism, breaking away from conventional Marxist-Leninist principles, and replaced pan-Kurdish aspirations by a dedication to the idea of a constitution of Kurdish rights within the principle of a ‘radical democracy’.14 After more than five years of extensive preparatory work, the Kurdistan Workers’ Party was eventually founded at a meeting that took place over just two days (26th and 27th of November 1978, now celebrated
The manifesto began with the first of its three chapters outlining ‘Class Society and the History of Colonialism’ (Sınflı Toplum ve Sömürgecilik Tarihi); analyzed ‘Kurdistan Society’ (Kurdistan Toplumu) in the second chapter—including a section, ‘The Turkish Republic’s colonization of Kurdistan’ (Türkiye Cumhuriyeti’nin Kürdistan’ı sömürgele tirmesi); and concluded by laying out the program for the ‘Kurdistan Revolution’ (Kurdistan Devrimi) in Chapter 3 (Öcalan 1978). 14 The initial aim of an independent Kurdistan and a revolution in the social structure of Kurdish society was adjusted in the early 1990s as the PKK abandoned the idea of independence, expressing instead the desire for a federal solution. Since 1999, after the capture of its leader Abdulah Öcalan and his imprisonment on an island in Turkey, the party has declared a unilateral ceasefire (in 2000), changed its name at its 8th Party Congress (in 2002) to Congress for Freedom and Democracy in Kurdistan, KADEK (Kurdish Kongreya Azadi û Demokrasiya/Kurdistan Kürdistan Demokratik ve Özgürlük Kongresi), and adopted a new political agenda on the idea of a ‘radical democracy’ and the establishment of a ‘democratic republic’ (implying a federal state or a semiautonomous region within Turkey). In this transformation, it has been argued, the PKK moved its focus from state building to subject-building, encouraging its followers in the development of a new homo-Kurdicus (Bozarslan 2004: 50). N.b. The party name was changed to Kongra-Gel in 2003 and back again to the PKK in 2005. In the period 2002–2005 the party was divided by internal struggle. The ‘reconstruction’ ( yeniden in a) of the PKK in 2005 symbolized the (re)establishment of the Abdullah Öcalan (‘İmralı’) federal line at the expense of what was called the ‘nationalist-separatist’ line represented by Osman Öcalan, Nizamettin Ta , Kani Yılmaz and others, who had separated from the party and established the Party of Democratic Patriots Kurdistan, PWD (Welatparêzên Demokratên Kurdistan Partîya) in 2004. Kongra-Gel was not abolished, but became effectively a continuation of the former ERNK, the name of the party body responsible for mass organization and politicization of the people in the region (‘. . . halkçı siyasetin yapıldığı ve halkın siyasal mücadelesinin yönetildiği demokratik kurumla masıdır’) (PKK 2005). 13
soldiers and settlements
55
as the party’s Founding Congress). A group of 25 people participated in this inaugural meeting, which established a Central Committee of seven persons headed by Abdullah Öcalan (Yüce 1999: 179).15 The meeting was organized in a small rural settlement known under its Kurdish name as Fis (Turkish name: Ziyaret) in the district of Lice in the province of Diyarbakır, the isolated location of the village and the house where the meeting took place providing excellent cover for the enthusiastic young men and women plotting to start a revolution in Kurdistan. The PKK planned to announce its existence with a spectacular action, the elimination of Mehmet Celal Bucak, a high-ranking member of the conservative Justice Party and exploitive landlord, who owned thousands of hectares of land with more than 20 villages and had the town and district of Siverek in the southeastern province of anlıurfa in his pocket. The planned assassination was not only a spectacular example of propaganda-of-the-deed to announce existence, but also revealed much about the PKK philosophy and modus operandi. Firstly, it was a declaration of war against the comprador, the landlord class collaborating with the Turkish state. The traditional Kurdish elite, of which Mehmet Celal Bucak was a member, had preferred to join in a patronage-based system with the Turkish State, and downplay its Kurdish identity. Participation in the political system gave this elite access to government power and resources. In return, the State acquired the means to impose indirect rule over Kurdistan and its people (Bozarslan, 1996: 141–42; Barkey & Fuller, 1998: 77; Bruinessen, 2003: 9). Secondly, the assassination of the landlord was intended to show a disaffected but diffident peasantry that there was an effective way to deal with the landlords and their grip over the people and the land—shoot them dead. However, the attempts to kill Bucak failed and backfired, pulling the PKK into a long war with his tribesmen.16 15 The Central Committee members were: Abdullah Öcalan, Cemil Bayık, ahin Dönmez, Mehmet Hayri Durmu , Mazlum Doğan, Mehmet Karasungur, and Baki Karer. 16 The nephew of Mehmet Celal Bucak, Sedat Bucak, was to become an important figure in Turkey’s ‘dirty war’ against the PKK. He established a militia of some 2,000 village guards (see below, 2.4), and became involved in special operations, as revealed by the Susurluk scandal (for the purposes of claiming state grants to support the village guards, he claimed to have 10,000 men in arms, which would imply a monthly cost to the State of $1.3 million, while on the HBB television channel, claimed to have 30,000 men!). The accidental crash of an armored Mercedes into a passing truck near the southwestern town of Susurluk on a cold night in November 1996 provided an indication as to the nature of the special operations units, generally supposed to be
56
chapter two
As the momentum of war developed, not just the Bucaks but others too, tribes and landowners, strengthened their ties with the state and fiercely resisted the guerrilla. In this, the State was quite willing to use tribes it viewed as criminal or delinquent;17 and, creating militia by enlisting their peasant populations as village guards, the landlords turned warlords also became an important element in the complex tapestry of local allegiances. From a strategic perspective, the elimination of landlords by the PKK had been intended as a first step, supposed to weaken the grip of feudal institutions over the land and people, and create room for maneuver for in the countryside—this giving early evidence of the PKK intention to organize a rural guerrilla (after the various failed attempts of the Turkish left to start a rural guerrilla, Kurdish resistance to the Turkish state was becoming increasingly metropolitan). Although the PKK would eventually start a prolonged people’s war from the countryside, the party first emerged from an urban and student environment. The foundation work for the PKK was done by Kurdish and Turkish students who had met each other in Ankara.18 The students disseminated propaganda, recruited members, and established regional committees (Kutschera 1994). In particular the Ankara section of the mass-movement, Revolutionary Youth (Dev-Genç), and the Ankara Democratic Higher Education Association, ADYÖD (Ankara Demokratik Yüksek Öğrenim Derneği )—both the product and context of activities of many leftist organizations—proved to be a fertile environment for the involved in and organizing death squads and drug trafficking. Along with Sedat Bucak (the only person other than the driver to escape the accident alive), the passengers in the Mercedes were Huseyin Kocadag, a top police official and the head of special teams; Abdullah Çatlı, an ultranationalist fugitive (convicted drug-trafficker charged with murder); and Gonca Us, Çatlı’s girlfriend (and wife by religious ceremony), a former beauty queen turned mafia hit-woman. (By the time of the accident, warlord Sedat Bucak had become an MP, representing anlıurfa for the DYP party, the then senior coalition partner and implicated in involvement with organized crime at the highest levels during its 1996–7 period in government under Tansu Çiller, along with the army, secret (MIT) and regular police forces, and various money-laundering business sectors (according to the Turkish media and official reports by the Turkish Government Inspector on Susurluk and the US State Department International Narcotics Control Strategy (INCSR); see Nezan 1988). 17 One of the most notorious was the Jirki tribe in Hakkari, whose chief Tahir Adiyaman was wanted for the killing of six gendarmes in 1975. Adiyaman made a bargain with the State, and, in return for absolution, raised a village guard force of Jirkis around Beytüsebap in the Serhat region (McDowall 2000). 18 Two housemates of Abdullah (Öcalan Kemal Pir form Gümü hane and Haki Karer from Ordu), celebrated in PKK circles for their leading roles in the formation of the PKK, were Turkish students. (Yüce 1999).
soldiers and settlements
57
recruitment of comrades in arms, and the political development and formation of the young revolutionaries. Their lengthy and frequent discussions contributed to the carving out of a distinctive ideology and the forging of kindred spirits (Yüce 1999). As early as 1972 these radical Kurdish and Turkish students had started to discuss the provocative thesis that Kurdistan was a colony and its inhabitants had the right to self-determination. This thesis was not attached to a plain nationalist approach, however. It was argued that liberation of Kurdistan was not an assignment for Kurdish nationalists, but for socialists, Kurds and Turks alike. In 1975 the group settled on a name, the Kurdistan Revolutionaries (Kurdistan Devrimcileri), but others knew them as Apocu, followers of Apo, the nickname of Abdullah Öcalan (apo is also Kurdish for uncle). The Kurdistan Revolutionaries did not consider the student and urban environment in Ankara to be well suited for the further advancement of their political and social struggle. They decided to disassociate from Ankara and establish themselves in Kurdistan.19 This decision was taken at a gathering in Ankara in 1976 (and what is called in PKK circles the ‘Dikmen-meeting’, named after the neighborhood where the gathering took place), and referred to as a ‘return’ (Yüce 1999: 261), even though the Turkish members of the group had never been in Kurdistan. The ‘return’ symbolized a retreat from urban areas, where state power was concentrated and strong, to rural areas, where state power was diffused and weak. Since it is easy for the state to concentrate power, including military might, in urban areas, these are tough environments for an insurgent strategy. Insurgency, the development of a counter-power and counter-institutions, including armed forces, was considered best organized in an environment where the State is absent or weak, i.e. in the countryside, in the köys and mahalles of the predominantly agrarian landscape of the Southeast. Insurgent Strategy Kurdish nationalism surfaced in the late 1960s and 1970s. Most of the political parties that emerged were either influenced by the Kurdish
19 The idea was to organize themselves in some of the larger cities and the surrounding areas in Kurdistan. Cemil Bayık went to Urfa, Kemal Pir to Mu , Haki Karer to Batman, and Ali Haydar Kaytan to Tunceli. A member of a rival organization called Sterka Sor (Red Star) killed Haki Karer in Antep in May 1977.
58
chapter two
nationalist movement in Iraq (Mustafa Barzani and the Kurdistan Democratic Party, KDP), or inspired by socialist/communist countries of the time, resulting in a rather fragmented political spectrum. Among the parties whose (ideological) origins could be traced back to the KDP were the Kurdistan National Liberators, KUK (Kürdistan Ulusal Kurtulusçulari), the Kurdistan Workers’ Party, KİP (Kürdistan I çi Party), a party dominated by landowners, despite its name), the Vanguard Party of Kurdistan Workers, PPKK (Partiya Pê eng a Karkerî Kurdistan), Rizgarî (Liberation) and Ala Rizgarî (Flag of Liberation). Other parties with some political influence among Kurdish intellectuals were KAWA, named after a famous figure in Kurdish mythology and inspired by Maoism and Albanian socialism, and the Socialist Party of Kurdistan, PSK (Partiya Socialîst Kurdistan), which was oriented towards the Soviet Union. The origins of both KAWA and PSK can be traced back to the Workers Party of Turkey, TİP (Türkiye İ çi Partisi), a classical communist (and parliamentary) party. The PKK had its origins in the new left movement in Turkey, in particularly Dev-Genç (Revolutionary Youth), inspired by experiences in Vietnam and Cuba. The PKK presented itself just as Marxist-Leninist, however, without any particular ideological affiliation to a specific ‘real socialist’ experience, a Soviet, Chinese, Albanian or Cuban model. The emergence of Kurdish nationalist organizations followed, ironically, the military coup of 1960. Initially, after a decade of political liberalization under Menderes’ Democratic Party, a narrowing of political space had been expected. General Cemal Gürsel, the leader of the military junta that seized power in May 1960, had boasted in November of that year that ‘If the mountain Turks [Kurds] give us no peace, the army will not hesitate to bombard and destroy their cities and villages. There will be a bloodbath of such dimensions that they and their country will no longer exist’ (PSK 1960). Yet the coup of 1960 was followed by a period in which restrictions were relaxed, eventuating in the establishment of organizations and journals giving voice, although carefully and shrouded, to Kurdish identity and aspirations. In Istanbul the bilingual journal Dicle-Fırat was established, and ran to eight issues in 1962–63 before being stopped.20 Other journals were
20 The journal editor, Musa Anter, was killed in Diyarbakır on September 20, 1992 by the JİTEM, a special (anti-terror) unit of the Gendarmerie.
soldiers and settlements
59
published, albeit with similarly brief lives (McDowall 2000: 404). The Labor Party of Turkey (TİP Türkiye İ çi Partisi), established in 1961, was also an important platform for articulating Kurdish aspirations. In 1968, following the full return of civilian government (parliament had been restored in 1961, but Gürsel had stayed on, as president, until 1966 the Revolutionary Eastern Cultural Hearths organization, DDKO (Devrimci Doğu Kültür Ocakları), was established, mainly active in the provinces of Diyarbakır and Batman. Another military coup occurred in 1970, however, and this was followed by a gradual increase repression of the Turkish left and Kurdish organizations. The decision taken at the Dikmen meeting, at which Abdullah Öcalan and other key-persons (the pre-PKK Apocu and Kurdistan Revolutionaries) decided to withdraw from Ankara and establish themselves in Kurdistan, was partly a reaction to this. Although repression was more severe in the main cities in the Kurdistan region in Turkey, in rural areas the state was almost absent. It was in a remote area in Turkish Kurdistan that the PKK was formally established as a political party in late 1978. With the formal establishment only predating the next military coup, in 1980, by a short period of time, the party was somewhat caught out in the process of party building. Nevertheless, it had already developed into an organization of some politically significance. It has been suggested that in 1980 the PKK was just one of the many Kurdish splinter parties, but records of arrests reveal that the PKK already developed in the major Kurdish political activist group. Turkish authorities arrested 3,177 alleged members and sympathizers of Kurdish organizations in 1980, more than half of which, 1,790 persons, were thought to be PKK related, and 667 were suspected members of the PSK and Teko în ( Jongerden 1997).21 The PKK militants who were not arrested kept a low profile in order to escape the junta’s dragnet. Abdullah Öcalan already had fled to Syria in 1979, almost a year before, his escape prompted by the State of Emergency which had come into effect in the Kurdish area the year before that, in 1978. In Syria, Öcalan continued to work on the political formation of the PKK. The first party congress was organized in Lebanon on June 15–26, 1981, one of the decisions taken being to 21 Of the remaining detainees, 459 were suspected KAWA members, and the other 1,261 allegedly members of different organizations (including the KUK, Rizgarî, Ala Rizgarî, and Riya Azadî/Özgürlük Yolu).
60
chapter two
withdraw the remaining party militants from Turkey, in order to save them from the ongoing military operations. The party also decided to prepare outside Turkey for guerrilla warfare. The PKK also regretted its conflict with other Kurdish groups (Teko în, KUK, Kawa, DDKD and others) at this congress, and later, in 1982, together with several parties from the Turkish left, the PKK established the Unified Resistance Front Against Fascism, the FKBDC (Fa izme Kar ı Birle ik Direni Cephesi) (Aslan 2005: 72–9).22 In Lebanon, the PKK found the infrastructure for guerrilla training. They settled in an abandoned training camp in the Beqaa Valley which had been used by the Democratic Front for the Liberation of Palestine (headed by Nayef Hawatmah), for the military education of party activists (Marcus 2007). Other leftist and nationalist Several Palestinian organizations—Ahmet Jibril’s PFLP-GC (Popular Front for the Liberation of Palestine-General Command) and George Habash’s PFLP (Popular Front for the Liberation of Palestine), Samir Ghosheh’s PPSF (Palestinian Popular Struggle Front), Yasir Arafat’s Fatah, and the Lebanese Communist Party—also provided assistance and extensive military training (Özdag 2003: 13; Yüce 1999; Marcus 2007). Preparing for a war against Turkey, the PKK first clashed with the Israeli army. Israel invaded Lebanon in June 1982, when the PKK was based in its training camps in the Beqaa valley. Forced by the situation, the PKK fought alongside Palestinian organizations against Israeli troops, which led to the death of twelve PKK guerrilla fighters (Yüce 1999). About a year later, in July 1983, a protocol was signed between the KDP (Kurdistan Democratic Party of Iraq) and the PKK, which prepared the ground for the establishment of PKK guerrilla bases in Northern Iraq, close to the international border with Turkey. In 1984, the party established a military wing under the name of the Kurdistan Liberation Union, HRK (Hêzên Rizgariya Kurdistan). The protocol between the KDP and PKK was terminated in 1987, although in truth relations between the two parties had deteriorated rapidly already by 1985, despite meetings between their leaders in Damascus, primarily as a result of PKK guerrilla attacks from Northern Iraq on targets in Turkey, and the resulting pressure of the Turkish military on the KDP to end the PKK presence in Northern Iraq (Kutschera 1994). At their 22 The parties participating were: the PKK, Dev Yol, TKEP (Türkiye Komünist Emek Partisi), TEP (Türkiye Emekçi Partisi), DEVRİMCİ SAVA , THKP/C-ACİLCİLER, SVP (Sosyalist Vatan Partisi) and TKP/İS (Türkiye Komünist Partisi/İ çinin Sesi). The FKBDC dissolved in 1986.
soldiers and settlements
61
second congress (this time held in Syria, over August 20th–25th, 1982) the PKK decided to initiate a prolonged guerrilla war from rural areas in southeast Turkey. Following the approach developed by Mao, the PKK envisaged a three-stage struggle: strategic defense (armed propaganda, small scale attacks, mobile warfare), followed by strategic balance, and then strategic offence (İmset 1995; Kutschera 1999; Özdağ 2003). The three-stage model involved a move from guerrilla to conventional warfare, with mobile warfare as transitional between the guerrilla strategy (based on the creation of a space to move) and the conventional (based on the creation of a space of control). In mobile warfare, the rules of regular warfare start to appear, but its guerrilla character remains. Large contingents of guerrillas concentrate not on fighting pitched battles, but on advancing deep behind enemy lines, attacking and then withdrawing swiftly (Özdağ 2003). The shift from guerrilla to mobile warfare was considered necessary to annihilate the enemy’s manpower and to liberate land. In combination with a popular uprising, the guerrilla army was supposed to force the Turkish army to leave all of Kurdistan (Özdag 2003), but this final battle was only to be started when the enemy was swamped, groggy, and demoralized, psychologically prepared for defeat (Pimlott 1985b: 59). Strategic Defense 1. Armed Propaganda. Aim: to gain a popular base of political influence, avoiding clashes with Turkish armed forces 2. Small Scale Attacks at unit level (8–12 guerrillas). Aim: to demoralize enemy and inflict losses 3. Mobile Warfare at level of battalion (40–50 guerrillas), regiment (110–120 guerrillas) and division (240–260 guerrillas). Aim: retreat of the army from the countryside into garrisons.
Strategic Balance Transition to tactics of conventional warfare. Increase of guerrilla forces from 4,000 to 14,000 armed men and women. Aim: hold and defend territories (liberated zones).
Strategic Offence Transition of guerrilla army into conventional army. Increase of the number of armed men and women to about 50,000. Aim: defeat of Turkish Armed Forces, liberation of Kurdistan
62
chapter two
Preparation for mobile warfare commenced in 1986 with a change in the military organization of the PKK and the replacement of the Kurdistan Liberation Union, HRK (Hêzên Rizgariya Kurdistan), with the Kurdistan People’s Liberation Army, AGRK (Artê a Rizgariya Gelê Kurdistan). The HRK had a guerrilla structure comparable with the Vietcong—the so-called ‘three-three’ structure in which three squads (of twelve guerrillas) constituted a team and three teams a company. The AGRK had an army structure comparable with a conventional army, with a hierarchy running up from a traditional platoon (40–50 combatants) to the level of a division (240–260 combatants). The transition from the HRK to AGRK reflected the PKK’s projection that the war would soon be entering the last phase of the first stage—the so-called mobile war—and marked the beginning of the use of the methods of conventional warfare. By 1990, the PKK had established hegemony in large parts of the Southeast, in particular in Serhat (covering parts of Hakkari, Van and some territories in Iraq and Iran), Garzan (covering parts of Batman, Siirt, Van, Hizan and Geva ), Botan (covering parts of ırnak, Hakkari, Van, Siirt, Eruh and parts of Northern Iraq), and Amed (covering parts of Diyarbakır, Bingöl, Genç, and Mu ). Although the PKK did not establish permanent control it succeeded in preventing the security forces from entering or remaining within these areas for long periods of time. These areas were considered to be one step away from the establishment of so-called ‘liberated zones’ and had the status of ‘semi-liberated zones’ (marking a transition from the phase of Strategic Defense to Strategic Balance). In many of the smaller settlements in the mountainous areas the PKK lodged guerrillas day and night. At times the PKK became overconfident—sometime during 1992, for example, the guerrilla commander ‘Dr. Baran’ (Müslüm Durgun) recklessly took his troops down the mountain and into the district town Ovacık in daytime. The first conventional battle took place between August 5th and 12th, 1991, when the Turkish Armed Forces attacked 24 PKK bases in Iraqi-Kurdistan. During this operation the ARGK resisted in positional warfare, defending territory, and did not dilute forces by ‘reverting’ to a war of movement. Despite suffering serious losses, the AGRK successfully defended its positions (Özdağ 2003: 39). Another indication for the transition to mobile warfare came from PKK attempts to extend control from rural areas to towns, occupying district towns such as Lice, Cizre, ırnak and Çukurca. Even though the PKK was only able to hold
soldiers and settlements
63
these towns temporarily and the military significance of the takeovers was limited, the political impact was tremendous. In June 1993, PKK leader Murat Karayılan confidently announced that the Parliament of Kurdistan would be established in the liberated area of Botan. It was now believed by many Kurds, and not only by those in sympathy with the PKK, that the PKK would succeed in reaching its goal.23 The Turkish Armed Forces Profile The Turkish Armed Forces, TSK (Türk Silahlı Kuvvetleri), are headed by four supreme command headquarters: the supreme commands of the land, the naval, and the air forces, and the gendarmerie. All but the naval forces have been involved in counter-insurgency operations. The air force flew combined reconnaissance-bombing missions (recce-bombing) against PKK guerrillas, both in Turkey and in Northern Iraq, but the bulk of the counter-insurgency work rested on the shoulders of the land force and the gendarmerie. The Turkish land force is by far the largest of the three service components. It divides the country into four military sectors on the basis of strategic conditions, terrain, logistics, communications, and the degree of potential (external) threat. The sectors are assigned to four field armies, employing a total of about 390,000 soldiers in 1991. The First Army, with headquarters in Istanbul, is widely deployed in the European part of Turkey, known historically as Thrace, with responsibility for the defense of that province, the Bosporus and Dardanelles straits, and the Kocaeli Peninsula. The Fourth Army is headquartered in Izmir and responsible for the vast area facing the Aegean coast, from the Dardanelles in the north to the southernmost Greek offshore islands. The First and Fourth Armies have defensive missions against Greece and the southwestern border of the ex-Soviet
23 Some men from one family, for example, who were closely linked to the PSK and not supportive of the PKK, bought two apartments in Diyarbakır in 1992. Living an extended settlement pattern of life in both İzmir and The Hague at the time, these men planned to settle in Diyarbakır, which they assumed would shortly be the capital of an independent Kurdistan. The apartments remained empty for years and were eventually sold in 2000 (author).
64
chapter two
block (Bulgaria).24 The Second Army with headquarters at Malatya and the Third Army with headquarters at Erzincan face the southeastern borders of Syria, Iraq, and Iran and the ex-Soviet Union (Georgia and Armenia). Both the Second and Third Armies were engaged in the war against the PKK, operating throughout the rugged mountains and deep valleys of southeastern Anatolia. In 1993 almost half of the Turkish land forces, about 185,000 soldiers, were deployed in the Southeast. The total number of forces involved in the war against the PKK, including gendarmerie, village guards and special units, amounted to somewhere between 300,000 (Human Rights Watch) and 365,000 (General Osman Pamukoğlu) (HRW 1995: 44; Pamukoğlu 2003).25 Regular policing roles are ascribed to the Turkish national police and the gendarmerie ( jandarma). The police employs around 120,000– 150,000 people and is primarily responsible for security in the urban areas. The gendarmerie is essentially a rural police force organized along military lines. In the 1990s, it comprised 70,000 personnel on active duty, and a further 50,000 reserves, with 40,000–50,000 men stationed in the Southeast. Responsible for security beyond the municipal boundaries of cities and provincial towns, as well as guarding Turkey’s international borders, the gendarmerie has jurisdiction over 90 percent of Turkish territory and 50 percent of its population. It is organized
24 Reference here to the ex-Soviet bloc reflects the traditional orientation of the Ottomans to their erstwhile Imperial Russian foes (there were a dozen Russo-Ottoman wars during the 250-year period of stagnation and decline of the Ottoman Empire, from its height in the mid-eighteenth century until Russian troops finally advanced well into northwest Anatolia prior to the 1917 revolution); and later, after a short period of rapprochement following the establishment of the Republic and the Soviet Union, the Turks’ European orientation saw them again at odds with their powerful northern neighbor, now with the USSR pit against NATO, of which Turkey had become a member in 1952. The period of the war with the PKK was also, of course, a time of uncertainty as the USSR unraveled and new countries were born. 25 The vast majority of these forces (80–95% of the army and gendarmerie, none of the special units, or village guards) would be temporary (male) conscripts on mandatory military service (askerlik), typically for an eighteen-month period, in what is effectively a rite of passage, both personal and social (family permission to marry and fulltime permanent employment are still commonly withheld until its completion). The occasion of military duty is widely held to be an event of celebration—at least by the less well educated and notwithstanding the hardships that will be endured (which may be considerable, especially, obviously, in time of war)—while the army in Turkey is generally held in high esteem by the people a whole. In terms of motivation, probably little can be read into the fact that most of soldiers are conscripted, but they are generally poorly trained, young and inexperienced (UHCR 2001).
soldiers and settlements
65
into thirteen regional commands encompassing Turkey’s 81 provinces. In each province, the principal gendarmerie commander, a colonel or lieutenant colonel, advises the governor on matters of security, and maintains direct charge of the district gendarmerie commands, usually headed by captains. Below the district commanders are commanders of the administrative sub-districts, each of whom controls the fixed posts in his area. There are some 3,600 posts, exclusive of border posts, usually located at intervals along the main roads and staffed by a sergeant and six or more gendarmes (Koivunen 2002: 149).26 On April 4, 1985, the then Prime Minister Turgut Özal added two articles to the Village Act (Köy Kanunu) which made it possible for the government to hire ‘Temporary and Voluntary Village Guards’ (Geçici ve Gönüllü Köy Korucuları) and legally sanctioned the state’s creation of an irregular paramilitary force. The Turkish armed forces effectively put into practice the institution of the village guard from 1987 onwards. Villages were expected to assign sufficient men to form a unit of village guards, which was armed, paid for and supervised by the local gendarmerie. The village guards were not only expected to take defensive positions against the PKK, but also to participate in operations, some of which involved cross-border incursions into Northern Iraq. About 5,000 men joined this paramilitary force in its first year, and by 1995 this number had increased to 67,000 (HRW-Arms Group 1995: 25). In 2003 the number of paramilitary was said to be about 59,000 strong (Table 2.1).27
26 Generally poorly educated and operating with near impunity in the countryside, the gendarmerie gathered a feared and hated reputation working with the Special Teams, committing extra-judicial killings and torture, and continue to be regularly cited in reports on human rights abuses (e.g. Amnesty 2004). 27 Probably many villages had to be coerced by the army into supplying village guards, whereas in others people joined up readily, perhaps as part of a landlord or tribal militia, as mentioned. Following the Susurluk revelations about Turkey’s ‘dirty war’, it should not be surprising if activities surrounding the village guard forces have been less than wholesome. The likelihood that they have been indirectly financed from state-administered drug trafficking is to be coupled with some of their own nefarious activities: a reported 23,000 village guards had been sacked for a variety of major crimes by 1997, the ‘divide and rule’ policy positively inviting the use of the war to frame the settling of unrelated disputes, feuds and vendettas. Also, regarded as traitors, Village Guards and their families were themselves the victims of some of the worst PKK atrocities. (DHKC 1997; Barkey & Fuller 1998: 147).
66
chapter two Table 2.1 Distribution of village guards in 2003
(Source: Göç-Der 2003: 10–11) Province Hakkari Van ırnak Diyarbakır Siirt Bitlis Mardin Batman Bingöl Mara Elazığ Mu Ağri Adiyaman Malatya Urfa ( anlıurfa) Kars Antep Tunceli Iğdır Ardahan Kilis Total
Number of Village Guards 7,643 7,365 6,835 5,274 4,680 3,796 3,360 2,943 2,533 2,267 2,115 1,918 1,881 1,510 1,392 966 578 565 386 374 96 34 58,511
Counter-insurgent strategy In spite of numerical superiority—in 1993, a state force of 185,000 (excluding the gendarmerie and village guards) pit against some 15,000–20,000 guerrillas—the army lost rapidly control. The Turkish military took up defensive and static positions, especially at night, when soldiers were thought to be safe in their enclosures. Garrisons were built and fortified, and army units confined themselves to these garrisons. Operations were carried out, but units returned to their barracks before dawn. The Turkish armed forces had decided to concentrate on the defense of larger settlements and to refrain from nocturnal operations, which gave the PKK considerable freedom to establish control in the smaller settlements and to move by night. Every now and then, the army would organize large sweeps, sending tens of thousands troops into an area, but these actions were not very effective as the guerrilla
soldiers and settlements
67
escaped into hiding while troops were being massed and during the operation, only to return after their retreat and dispersal. In warfare theory, the conclusion of counter-insurgency specialists is that sweeps don’t work (Tomes 2004). American military sources ascribed the bad performance of the Turkish army during the 1986–1993 period to the lack of an integrated strategic counter-insurgency doctrine (SOT 1997). Until the dissolution of the Warsaw Pact in 1990, the main mission of the army had been that of a static defense, aimed at countering Soviet and Warsaw Pact forces in the Caucasus and any possible attack on Thrace. The PKK posed a very different threat. Over time, and at some cost, the army learned that a counter-insurgency war could not be fought from a defensive and static position. There were serious deficiencies in the army’s ability to respond with speed and flexibility, deficiencies that became even more clear with the spread of guerrilla warfare over the entire region of southeastern Turkey. The Turkish armed forces thus formulated an integrated doctrine of area control, named the ‘field domination doctrine’, and aimed at the production of a new (contracted rural and urban) war space. The doctrine had been laid down in 1991, but not put into practice until after the reorganization of the army, initiated in 1992, was completed in 1993. The PKK was recognized as the first priority threat, and a change was announced from ‘search and destroy’ sweeps to a ‘clear and hold’ penetration strategy (Özdag 2003). The origin of these two strategies goes back to a fundamental disagreement between the Americans and the British during the Vietnam War. The British mission in Vietnam had developed a counter-insurgency plan in which regular forces would clear and take hold of an area. The Americans, however, preferred a ‘search and destroy’ role for its troops—but eventually came to realize that it was very difficult to find the enemy in large-unit sweeps (Dunn 1985). The Turkish change from ‘search and destroy’ to ‘keep and hold’ marked an important shift from an ‘American’ to ‘British’ style of counter-insurgency. The ‘field domination doctrine’ implied, among other things, that the armed forces would abandon the approach of garrison-line-ofdefense. ‘Garrisons,’ wrote General Osman Pamukoğlu, commander of the Hakkari Mountain Warfare and Commando Brigade between 1993 and 1995, ‘do not provide protection, but because of their static disposition are targets for the enemy.’ Commando brigades and Special Forces became key elements in the war against the PKK. Their
68
chapter two
soldiers were supposed to stay in the field day and night, searching for PKK units and camps. It is ironic that by 1993 the army had started to apply guerrilla tactics against the PKK at a time that the PKK was preparing for conventional warfare. The war became not only one of movement—numerical superiority was also considered a key element in the establishment of area control. Between 1993 and 1995, the number of Turkish troops in the region was increased from 185,000 to 360,000 (including as many as 70,000 village guards). The majority of forces continued to perform regular duties. In Hakkari, for example, the land forces comprised fourteen battalions amounting to 56,000 soldiers in 1995, but only five battalions were employed in a war of movement. These five battalions were organized into special units of ‘go-getters’ who lived in the mountains for weeks, hunting the guerrilla. The nine other battalions had regular tasks to fulfill, generally involved in patrolling an assigned area. When PKK guerrillas were spotted, their mission was not to engage in contact, but to surround them. Special units were transported to the area by helicopter and took up the pursuit. Following the new doctrine, the General Staff decided to reorganize the army and shifted from a relatively cumbrous divisional and regimental structure (designed in the period of the Cold War) to a relatively flexible corps and brigades structure (to fight a multi-directional and multi-dimensional internal enemy), which was supposed to contribute to a more rapid response and higher mobility (Ministry of Defense 2000). Prior to the army reorganization, the principal tactical units had consisted of sixteen infantry divisions and one armored division, plus twenty-three independent brigades, of which six were armored and four mechanized. Basically, the Turkish land force was a large but badly equipped infantry force. Under the reorganization, all except three of the seventeen infantry divisions were dismantled. The existing nine corps were retained, with brigades directly responsible to the corps commands. Parallel to the sweeping 1992 reorganization of the army, several other significant changes and additions were made to the structure of the Turkish land forces. Most importantly, there was a major strategic shift, with the army being handed the lead role in the fight against the PKK. Before 1992, the gendarmerie had been assigned the job of fighting the guerrilla, but it was ill designed to perform such a task. The Turkish gendarmerie lacked hardware and had relatively few officers (commissioned or non-commissioned). The main burden of counter-insurgency
soldiers and settlements
69
Table 2.2 Structure of Turkish army principal tactical units
before and after 1992 reorganization Divisions Before After Infantry Armored Mechanized Commando Total
16 1 17
Brigades Before After 13 6 4
3
23
9 141 172 43 444
1 One armored brigade composed of 6 battalions—2 armored, 2 mechanized and 2 artillery. 2 One mechanized brigade composed of 4 battalions—2 mechanized, 1 armored, and 1 artillery (+1 reconnaissance squadron). 3 One commando brigade composed of 4 battalions—3 commando and 1 artillery. 4 Not including presidential guard (1 brigade), border defense (5 brigades), coastal defense (1 brigade) and the marines (1 brigade), all counted as ground forces (IISS 2005).
had thus fallen on a poorly equipped force of poorly trained conscripts lacking professionalism or experience.28 With the reorganization of the army specifically in order to deal with the internal guerilla threat, the gendarmerie was able to take a back seat and assume a more supportive role. (Özdag 2003: 2). Also, from 1993 onwards, the gendarmerie was itself tailored to fight the PKK. Specialized sergeants were recruited to replace ordinary conscripts in local, low-level leadership positions, and the gendarmerie was also made better equipped, with additions including helicopters, crucial in a war of movement against guerrilla forces (GlobalSecurity.org 2005). Similar developments within the army also were enacted, with an upgrading of personnel infrastructure and hardware. First, the ratio of personnel with actual experience in irregular warfare was increased through its so-called ‘sergeants with tenure’ program (Özdag 2003: 53). Also the armed forces improved the network of roads and communication facilities, and modernized its armaments and means of transport for more effective employment of troops over land and by air. The Turkish armed forces benefited from the Treaty on Conventional Armed Forces
28 In contrast to France, Belgium and Italy, where operations by the gendarmerie are carried out mainly by officers, in Turkey they are not, and it is conscripts who form the bulk of the service (see note 24 above).
70
chapter two
in Europe (CFE Treaty) signed in Paris on November 19, 1990, under which provisions NATO and former Warsaw Pact countries agreed to reduce the size of their conventional forces and conventional armaments. As a result of the treaty, Turkey was able to obtain armored infantry fighting vehicles and armored personnel carrier vehicles, from both NATO countries and Russia. Additionally, Turkey obtained modern troop-lift helicopters from the United States, enabling a more effective deployment of commando and infantry troops. Another major change involved special units. Existing special units were deployed, both army units and others. First, elite army units (the Bolu and Kayseri brigades) became heavily involved in counter-insurgency operations in the Southeast, as did army special forces, about which relatively little is known, and special operations police and gendarmerie teams.29 Second, Special Action Teams (Özel Haraket Timleri), under the authority of the police, and Special Teams (Özel Tim), under the authority of the gendarmerie (these with a combined total of 15,000–20,000 men), were created, which, along with the anti-terror and intelligence department of the gendarmerie, JİTEM ( Jandarma İstihbarat ve Terörle Mücadele Te kilatı), operated as death-squads, identifying and killing alleged PKK cadres. Finally, the previously poor performance of state forces was also identified as low levels of information about the enemy, so the Turkish armed forces also tried to improve their intelligence. Intelligence as to the whereabouts of the PKK was often contradictory, wrong, or incomplete. The population did not provide the authorities with information about the hideouts of PKK units, the identity of PKK fighters, or coming operations. In order to secure timely and accurate intelligence, it was emphasized that the army had to show it was in control. The military thought that one of the reasons that people were more inclined to give information to the PKK than to the Turkish authorities was that the PKK had more control of the area. More army control over the population was supposed to result in more intelligence 29 The special operations police teams were organized in 1983 under the Special Operations Branch Directorate (Özel Harekat ube Müdürlüğü) and Special Operations Group Authority (Özel Harekat Grup Amirliği) in the cities of Ankara, İstanbul and İzmir, for the purposes of undertaking ‘high risk’ ‘counter-terrorist’ operations. After 1993, these special teams were promoted to a position of active combat against the PKK. They operated in teams of at least twenty, upon the direction of military units under the responisbility of the army if outside a police region (Be e 2006: 118–120).
soldiers and settlements
71
(Pamukoglu 2003: 58). The authorities also started to take an interest in social research, turning the people and the region in objects of study in order to know them better (Özök 2004). International Perspectives Like the American counter-insurgency specialists in the 1970s (Halperin 1974: 288–89), Turkish generals involved in the war against the PKK started to study relevant manuals and works of military theory in order to acquaint themselves with previous histories and the basic principles of guerrilla warfare. These included guerrilla textbooks—e.g. Mao Zedong’s ‘Basic Tactics of the People’s War’, Alberto Bayo’s ‘What is Guerrilla?’, and Carlos Marighella’s ‘Handbook of the Urban Guerrilla’)—as well as Turkish army instruction books from 1934–1936 (Pamukoğlu 2003: 61–2). The content of these Turkish army manuals is not certain, but it is known that several secret documents -in the 90s had been -in the 30s written in relation to the pacification of rebellious areas in the Kurdistan region in Turkey during the thirties. In relation to the Dersim area, envoys now advised the government in secret reports to destroy those rural settlements in areas i) difficult to access, and ii) at strategic locations, in order to make possible an effective control of people and peoples’ movements (Anonymous 1998: 199). Construction of new roads was recommended in order to afford government forces better access to the area. The total or partial deportation of the population was advised concomitant with a dispersed settlement of their populations in ‘real Turkish’ villages (ibid. 182–183: Bayrak 1993: 506–509). In addition to revolutionary handbooks and old army manuals, army textbooks on counter-insurgency operations in (non specified) ‘third countries’ were studied (Pamukoğlu 2003). There are indications that American specialists and Vietnam veterans instructed special forces of the Turkish army in the effective combat of Kurdish insurgents ( ahan & Balık 2004). Little is known about American involvement, but we do know that American specialists and CIA intelligence officers were already instructing Turkish forces and had been since the 1970s, when the American counter-insurgency specialists had instructed special units of the Turkish Army to combat leftist militant and guerrilla organizations during the 1970s. According to former Turkish military prosecutor and Supreme Court judge Justice Emin Değer, there were close ties between the CIA and Turkish army
72
chapter two
counter-insurgency units (Değer 1977). It has also been argued that the US military mission was housed in the same building that headquartered the Turkish contra-guerrilla department (Roth and Taylan 1981). Even though Vietnam and the American involvement (and ultimate failure) there provides the central image of guerrilla and contra-guerrilla warfare, the most successful example of counter-insurgency was actually that of British troops in Malaya in the 1950s (the British military mission in Saigon also advised the Americans in Vietnam, advice that was initially ignored and only implemented when it was already too late). The combat of British troops against the communist guerrilla in Malaya (1948–1955) is regarded by military experts as the piece de resistance in counter-insurgency, a model operation and one of the greatest successes ever in anti-guerrilla warfare (Newsinger 2002: 31). Basically, the Malayan situation resembled the Turkish one in many respects. The Malayan Communist Party’s guerrilla strategy involved widespread and continuous attacks on forces associated with the British, attacks that were supposed to result in the British concentrating their forces on the defense of communications, supply lines and towns, and thereby allowing the guerrilla to then switch to conventional warfare tactics in order to liberate rural areas, establish administration, and build up forces; as their strength increased, they would later expand the liberated areas and launch large scale attacks, until the British position would become untenable and their forces compelled to surrender and leave. Essentially, this was Maoist guerilla-mobile-conventional warfare transition strategy—the PKK had a similar strategy in mind, and the Turkish armed forces applied against them many of the same British tactics which had proved so successful in Malay. First, the British had brought in overwhelming force. They increased the regular police from 10,000 to 40,000 and including auxiliaries numbers reached a peak strength of 100,000 in 1951 (Newsinger 2002: 46). The Turkish armed forces similarly moved in more troops, with the proportion soldiers to guerrilla increasing from 30 to 1 in 1993 to 60 to 1 by 1995 (Özdağ 2003). Second, the British had moved in combatants to Malay who had gained experience fighting (albeit losing) an irregular war against the Zionist insurgency in Palestine and enlisted them as sergeants in the army that fought against the Malayan communists (Newsinger 2002: 46). Likewise, the Turkish armed forces started its so-called ‘sergeants with tenure’ program to increase the ratio of experienced soldiers (Özdağ 2003).
soldiers and settlements
73
Third, numerical superiority and increased experience had been combined with a painstaking approach involving small patrols and ambushes, guided by improved intelligence. These tactics replaced a failing broad-brushed approach that involved sweeps over large areas where the guerillas were expected to be hiding (Newsinger 2002: 46). The Turkish armed forces made the same transition in adopting the ‘clear and hold’ strategy (Özdağ 2003). Fourth, the British had established in Malaya an indigenous armed force, the Home Guard. Complete populations of villages were drafted into paramilitary militias. The establishment of a local, indigenous militia enabled regular troops to be freed from defensive positions and redeployed more aggressively in the anti-guerrilla war (Marston 1979: 46). In Kenya also, the British had formed so-called village guard detachments, which freed regular forces from static defense roles (Pimlott 1985a: 22), and in Vietnam the Americans too established paramilitary forces, deployed at platoon strength in villages and hamlets on their arrival (Dunn 1985: 94). Again, the Turkish armed forces did likewise, supporting the establishment of an indigenous militia of village guards manned mainly by Kurds.30 Fifth, the British counter-insurgency forces in Malaya had imposed rigorous food controls and resettled populations away from areas of guerrilla activity. Between 1950 and 1952, the British resettled about 400,000 people, mainly Chinese, in 400 new villages (Newsinger 2002: 50).31 Resettlement of the population was intended to isolate the guerrilla forces from the population and deprive them of finance, intelligence, food, shelter, and recruits, while at the same time creating kill-zones in the open and empty countryside. The Malayan communists reacted by retreating into enclaves deep in the countryside, in mountainous areas where they had to devote their energy to mere survival, at the expense of offensive action (Pimlott 1985a: 22–23), and where they could be pursued eventually by the armed forces of the state (Marston 1979: 49).32 30 Only the French in Algeria had reservations about arming local militias (Pimlott 1985b: 57). 31 The British had also attempted to resettle the indigenous Malayan populations, but this was abandoned after between a quarter to a third of the resettled aboriginals native inhabitants died of disease and despair in these camps (Newsinger 2002: 55). 32 The reason for success of the British military counter-insurgency was also related to two important problems the guerrilla had to cope with. First, the guerrilla had a ‘Chinese’ character and never really gained extensive support among the majority
74
chapter two
The British resettlement operations in Malaya were carried out with force and brutality. In the course of the resettlement, houses and crops were burned, agricultural tools smashed, and livestock killed. The newly created settlements were little more than camps surrounded by a barbed wire fence and illuminated by searchlights (Newsinger 2002: 31–59). Clearance of the countryside satisfied a military need, but the prisonlike camps were not really appropriate environments to win the ‘hearts and minds’ of the people. It was argued that not only was evacuation a military necessity, but also that resettlement provided an opportunity for the state to (re)establish its relationship with the population. The new settlements, therefore, were to be developed as starting points for development, and also, importantly to be perceived as such (Sutton 1981; Sutton and Lawless 1978; Zasloff 1962–1963). In South Vietnam, the Americans took up the idea of disassociating resettlement from narrow military objectives and incorporating into resettlement the idea of development. This eventuated in a plan for large-scale rural resettlement in the southern part of South Vietnam, announced in 1959 and implemented with some urgency over the subsequent two years. Essentially, the scheme sought to regroup peasants into rural concentration centers, termed ‘agrovilles’. A compelling reason for the establishment of these centers was security, as outlined by the then Vietnamese Minister of Agriculture in an instruction to governors, the prefects of Saigon and Cholon, and the mayors of all large cities: The reason for this work is that the population, especially in the South, is living in such a spread out manner that the government cannot protect them and they are obliged to furnish supplies to the Vietcong. Therefore, it is necessary to concentrate this population . . . (Zasloff 1962–1963: 328).
The new, nucleated settlements were expected to bring the population living in small and dispersed settlements nearer to roads and the main arteries of communication, enabling military forces to provide surveil-
Malay population. The British established positive political links with non-communist Malayan political parties, and gradually moved towards independence for Malaya, which the British made conditional on crushing the communists. In the end, the guerrilla was not fighting British colonial rule, but the new government of an independent Malaya. Second, the guerrilla relied completely for their resources on whatever they could gain locally—they had no bases in other countries, nor did they ever receive substantial support from the outside world (Newsinger 2002).
soldiers and settlements
75
lance and control population movements more easily. It was hoped that in these ‘agrovilles’ the population could be isolated from the guerrilla. The agrovilles were not single settlements, but compact conglomerates, consisting of a central settlement and one or more dependent settlements. Eighty center-villages were envisaged and an unknown number of dependent villages. The head of the British Advisory Mission in Vietnam, Robert Thompson, who had helped implement counter-insurgency measures in Malaya in the 1950s, proposed that the South Vietnamese integrate economic and social programs into their resettlement policy in order to (re)establish influence in the heavily populated Mekong Delta. Resettlement was to be disassociated from exclusive military concerns, and coupled to development targets. The then president of Vietnam, Diem stated: This year I propose to create densely populated settlement areas in the countryside, where conditions are favorable to communication and sanitation and where minimum facilities for the grouping of the farmers living in isolation and destitution in the back country exist. These settlement areas will not only improve the life of the rural population, but they will also constitute the economic units which will play an important role in the future development of the country as a whole (ibid.: 327).
The agroville was portrayed as the happy marriage of a bustling city life and placid rural existence (its other name was ville charnière or ‘hinge city’). Just years after its announcement, however, the idea of agrovilles was abandoned, to be replaced by ‘strategic hamlets’, also known as ‘new life hamlets’ (Donnell 1970). The strategic hamlets were smaller in size, and would not require the massive resettlement of an unwilling population. They were meant to aggregate into a fine-meshed network of locations under control of the central government. The development of a bureaucracy and the organization of a local political party system was to contribute to ‘state-building’ in the countryside (the only functioning institution for the integration of the population under the state was a local militia force).33 Further instructive parallels to the Turkish case can be found in Algeria. Like Turkey, the rural settlement structure in Algeria was
33 Like the agroville policy, however, the strategic or new life hamlet policy also failed to materialize and the rural population never was resettled.
76
chapter two
dispersed and lacking nucleated settlements. This settlement system underwent considerable upheaval between 1954 and 1961 with a mass eviction of the rural population and the concentration of at least two million people into about 2,300 so-called ‘regroupment centers’ (Centres de Regroupement) (Sutton 1981). The French colonial authorities used the policy of concentrating population to destroy resistance, weaken social (tribal) structures, and colonize fertile land for European settlers. Between 1955 and 1957, rural population concentration was the result of a policy creating no-go zones, particularly in mountainous areas, no-go zones that became free-fire zones for the French ground and air forces in their struggle against the insurgent National Liberation Front, NFL (Front de Libération Nationale). Initially, no provision was made to resettle the displaced population, who fled to villages and towns in the foothills of the mountains and in the plains. The Turkish resettlement of population in the Kurdistan region in the 1990s resembles this initial policy of the French. It was only later, after 1957, that the French created the regroupment centers. This policy was not a default to the creation of no-go zones, but an orderly complement (Sutton and Lawless 1978). The establishment of regroupment centers was a subject of debate and suffered from discord between the military and civil authorities. The military created temporary regroupment centers with the aim of clearing territory to their rear, without giving any real consideration to their future development. The civil authorities, however, sought to transform the consequences of military operations into a belated attempt at rural development, and wanted to create permanent new settlements only where they might develop into viable agricultural communities (ibid.: 333). The ‘new villages’—as the regroupment centers were termed—were to be provided with basic amenities. Nevertheless, by 1961 there were still more temporary ‘military’ settlements than ‘civilian’ development centers.34
34 Of a total of 2,380 centers 1,217 were temporary regroupment centers and 1,163 permanent ‘new villages’ (Sutton and Lawless 1978: 331–350).
soldiers and settlements
77
Box 2.1 Resettlement in Algeria In some cases, concentration of the population in new settlements involved a movement of only 3–5 km. In others, the people had to move to more distant localities. New settlements were usually located near a road, which guaranteed good access for the colonial forces. Even if the population movement was geographically limited, its impact was enormous. Standing agricultural plots were generally located in no-go zones. Although in some occasions the rural inhabitants were allowed to cultivate their land, the pattern of work had to conform to a military and not an agricultural timetable. As a result, agricultural work was neglected on a large scale. Crop production dropped and most livestock was lost. Additionally, forests that had provided complementary resources for the local people were now closed and in a number of cases affected by napalm bombing. Alternative sources of income to maintain the economics of day-to-day life were scarce. Where possible, peasants hired out themselves to work on French estates or engaged in trading activities, and where this was not possible, they joined the army of the unemployed. The morphology—shape and structure—of the regroupment centers and new villages conformed to the military requirements of control and supervision. No longer were dwellings dispersed and loosely clustered as in the mountain settlements: the new settlements were compact and geometrical constructions, sometimes with a simple grid-iron lay-out. The houses were built from concrete and corrugated iron, and consisted of two rooms, a kitchen and a small courtyard. Each settlement was usually surrounded by barbed wire fences and overlooked by one or more watch towers. The new settlements not only concentrated, but also compartmentalized the population, not just because of their individual morphological structures, but also because a failure to establish good linkages between the new settlements. At independence in 1962, it was expected that many of the displaced persons would abandon the forced settlements and rebuild their devastated hamlets and villages. However, only a minority of about 25 percent (some 250,000 persons) actually did so. Return was most limited in the plains and in areas in close proximity to towns and roads, but it was still significant in the mountainous areas. According to Sutton and Lawless the disruption in social life produced by the resettlement had left many peasants in a limbo, caught somewhere between a rural and urban life, yet neither one nor the other. The displaced were no longer peasants, but not yet urban dwellers either (Sutton and Lawless 1978; Sutton 1981).
78
chapter two Environment deprivation
The armed forces were quite suspicious of the villagers in so-called ‘PKK populated areas’ (Özdag 2003: 33), and enforced firm measures against the rural populations. In addition to curfews, villages were put under food-embargoes, allowed only to bring limited amounts of flour, rice and other food products into their settlements. The armed forces started to evacuate and destroy villages considered supportive for the PKK. This came to be increasingly regarded as a productive strategy, since it cut off the guerrilla from the intelligence, shelter, recruits, and food supplies, on which it depended (in Mao’s terms: ‘Because guerrilla warfare basically derives from the masses and is supported by them, it can neither exist nor flourish if it separates itself from their sympathies and co-operation’) (Asprey 2002: 255). The military openly hinted at large scale evacuation of rural settlements and the concentration of the population in a few large settlements (e.g. see quotation from General Osman Pamukoglu at chapter head). The evacuation of rural settlements was made possible in 1987 by the Decree No. 285 concerning the State of Emergency, which granted the governor formal authority to order the temporary or permanent evacuation of settlements in the interest of ‘public security’ (in real terms military commanders decided about the evacuation of rural settlements). Governors in State of Emergency provinces hinted at the concentration of the population in centers where they could be controlled effectively (see Governors’ Letters in Oyan 2001a). However, resettlement of the population was not declared openly as official (civil or military) policy, since the ratification of the 2nd Protocol Additional to the Geneva Conventions had banned this practice in 1977. The evacuation of villages was not a side-effect of the counterinsurgency of the Turkish Armed Forces—notwithstanding the gradual and undeclared way in which it was introduced and extended—but one of its primary aims, intended to contribute directly to the ‘environmental deprivation’ of the guerrilla. Rural guerrilla fighters prefer to operate in areas providing both plenty of cover and concealment (typically mountainous and forested areas) and also practical support from the indigenous population—a physical and social environment that operates as a force-multiplier for the force conducting a war of movement (the guerrilla). For the force holding a static position of territorial occupation (the state), areas where the population is dispersed, the physical terrain rugged and the communication and transport
soldiers and settlements
79
infrastructure poor are difficult to command. They have the effect of a force-diffuser (the military ‘dilute’ in the territory). This is a serious disadvantage for supervision and control of the population and effective combat against the guerrilla. The evacuation of villages was a means of destroying these positive environments for the guerilla, which were negative environments for the state. It was intended to force the guerilla either into isolated retreat (high in the mountains, across the border with Iraq) or else into undesired or over-hastily planned combat in ‘urban’ environments.35 The doctrine of ‘field domination’ was translated into what was effectively a policy of ‘draft or destroy’. Villagers were placed before a frightening dilemma: they could either become members of the paramilitary village guards and take up arms against the PKK, or be evacuated and see their houses and villages destroyed. The evacuation of villages was supposed to destroy the social environment of the PKK. The countryside, once the livelihood of the insurgents, was to be altered in a kill-zone and the insurgents forced to retreat high into the mountains or cross the border to Iraq, where they were relatively safe, but suffered from hardship and isolation from the rural population, their source of supplies. The ‘paramilitarization’ of the peasant population was to create a permanent state of low-level counterinsurgency in the preferred physical environment of the PKK (while also releasing the army from the impossible task of having to garrison every single settlement). The evacuation of the rural settlements was no collateral damage or simple reprisal, but a constituent part of the counter-insurgency program. Evacuation took place on a large scale. Since reliable statistics are not available, the numbers of internally displaced persons and evacuated or destroyed villages are necessarily estimates. Human rights organizations have claimed the figures issued by the authorities to be of doubtful provenance and contradictory (HRW 2002: 25). According to official figures from 2001, for example, 55,606 households, or 384,792 people were evacuated through the decade of the 1990s, from 833 villages 35 The evacuation of villages did, in fact, lead to a contracted environment for guerrilla warfare (itself then penetrated by special units using guerrilla tactics) and a retreat of PKK forces. The PKK did not switch to urban guerrilla warfare, but it did establish a ‘front’ with the Revolutionary People’s Liberation Party/Front, the DHKP/C (Devrimci Halk Kurtulu Partisi/Cephesi), a Turkish, left-wing, urban guerrilla. The attempted collaboration proved ineffective, however, and was terminated by the DHKP/C in 1998.
80
chapter two
and 2,382 small rural settlements (totaling 3,215 settlements) in fourteen provinces in the East and Southeast (Adiyaman, Ağrı, Batman, Bingöl, Bitlis, Diyarbakır, Elazığ, Hakkari, Mardin, Mu , Siirt, ırnak, Tunceli and Van: see Graph 2.1) (Oyan et al. 2001a). Ministry of Internal Affairs figures from February 2006, however, put the number of villages evacuated at 945 and the number of hamlets (referred to as sub-villages) at 2,021 (totaling 2,967—248 less than five years earlier!) (Dağ 2006).36 Notwithstanding the mysterious reduction in settlements affected, the population figures rose: the 2006 figures determined the number of households affected by the evacuations at 58,856 (42,220 from villages and 16,639 from hamlets), and the number of persons displaced at 385,335 (245,605 from villages and 112,730 from hamlets). Examining this rise more closely, we see that the additional evacuees between 2001 and 2006 supposedly numbered 542 people from 3,250 different households! The approximate number of settlements depopulated and destroyed (about 3,000) is not really in dispute, but the number of people affected has been a subject of great controversy. Human rights organizations and NGOs claim that Turkey deliberately presents low numbers to camouflage the magnitude of the displacement. Some estimate the number of displaced persons to be as high as three or even four million (KHRP 2002). Other NGOs, however, such as the Turkish Economic and Social Studies Foundation, TESEV (Türkiye Ekonomik ve Sosyal Etüdler Vakfı), consider the number of three to four million displaced persons to be a rather high estimate, and tend more towards 1.5 million (Aker et al. 2005: 8). Research conducted by the Hacettepe University Institute of Population Studies under the coordination of the Prime Ministry’s State Planning Organization, DPT (Devlet Planlama Te kilat) puts the figure at between 950,000 and 1,200,000 (Tezcan & Koç 2006).37
36 Figures presented by the Ministry of Internal Affairs, Strategy Development Board at ‘The Internally Displaced Persons (IDP) Conference’, Ankara, February 23, 2006, co-organized by the Turkish Ministry of Interior Affairs and the United Nations Development Program (UNDP 2006). 37 The Hacattepe survey was carried out in the fourteen provinces focused on by Oyan et al. (2001a), between December 2004 and June 2006 (following a 2002 UN recommendation), and covers people displaced during the period 1986–2005. It gives (rather specific) estimated figures of between 953,680 and 1,201,200 IDPs (‘securityrelated migration’ within and out of the fourteen provinces), 80% of whom originated from rural areas and 20% form urban.
(Source: Kurban et al. 2006)
Graph 2.1 Internal displacement in Turkey
soldiers and settlements 81
82
chapter two
Numbers provided by the IHD suggest that most evacuations occurred in the period 1991–95, peaking in 1993–94 (see Table 2.3). The KHRP report divides the process into five main sections: the Initial Period (Ba langıç Dönemi), 1985–89; Centralization of Village Evacuation (Köy Bo altmalarının Merkezile tirilmesi), 1990–91; Systematic Village Evacuations (Sistematik Köy Bo altmalar), 1992–93; Escalation of Village Evacuation (Köy Bo altmalarında Hızlanmı ), 1994; and then the (unnamed) year groupings of 1995–96, 1997–99, and 2000–01 (KHRP 2002). Table 2.3 Number of evacuated and burned villages 1991–2001
(Sources: İHD 2001; KHRP 2002) 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 109
295
874
1,531
243
68
23
30
30
–
3
The Turkish Armed Forces preferred settlement of the evacuated population in urban entities, where the concentration of counter-insurgency units operated as a force-multiplier. The village evacuations did indeed effect a sizable demographic shift from rural areas (not under the control of the Turkish Armed Forces) to urban areas (relatively well controlled by Turkish security forces) as the people left (and fled) their villages and hamlets and made for the towns and cities. This population movement was not schemed as such, and resulted in huge numbers of displaced persons indefinitely housed in unprepared urban environments (Box 2.2). From a military perspective, this disregard of resettlement schemes was a logical and rational choice. Empirically, the case against resettlement is overwhelming. The schemed evacuation of villagers has been part of only one successful counter-insurgency campaign (that of the British in Malaya), and military experts warned that the success of resettlement in Malaya should be regarded as an aberration rather than replicable model (Marston 1979: 49). Box 2.2 Schemes The aim of the resettlement was military: environment contraction and deprivation of the guerrilla of its environment. Resettlement took the form of non-schemed, forced migration, creating resettlement tracks from rural to urban areas. Only with the involvement of civil authorities in 1994 did the idea of resettlement schemes enter the political agenda. Several plans were made to scheme the resettlement of rural populations in urban-type
soldiers and settlements
83
settlements, both near cities and in the countryside, but these plans largely failed to materialize. Two prominent examples of these urban-type resettlement plans were the 1994 toplu kondu and toplu çiftlik plans. Developed by the Housing Development Administration of Turkey (Toplu Konut İdaresi Ba kanlığı), the toplu kondu (collective shelter) project was designed as a hybrid of the shantytown type dwellings known as gece kondu (literally night shelters, implying that they are put up overnight, under cover of darkness), and the government housing program called toplu konut (collective housing). In the toplu kondu blocs the government would provide for a basic, one-floor shelter of 538 square feet built on a piece of land that varying in size between 2,100 and 4,300 square feet. Inhabitants were supposed to construct additional floors, extra rooms or workplaces according to their own needs and means. Several of these toplu kondu blocs, with a total of 8,000 shelters, were planned far away from the depopulated areas, in rural the areas near the economic growth-poles of Adana, Urfa, Diyarbakır and Gaziantep. Another 2,000 shelters were planned in the proximity of several district towns in these areas. Turkey requested for a loan of $50 million from the World Bank to prime the scheme. After initially favorable moves, however, the World Bank pulled out, for reasons that were never explained. The Turkish authorities did not decide to finance the schemes from their own funds (Meclis Ara tırma Raporu 1997: 44). A similar plan proposed by then Prime Minister Tansu Çiller envisioned the construction of large settlements with an average population of 1,000 inhabitants apiece. These settlements were referred to as toplu çiftlik modeli projesi (collective farm model projects), or simply toplu çiftlik (collective farms). Despite its name, the toplu çiftlik plan was meant to resettle the population in the vicinity of urban areas. According to a government circular dated October 24, 1994, the resettled people would again be provided with shelters of 538 square feet, but now were also to be given access to agricultural land (50 acres of irrigated land or 120 of rain-fed land) (Genelge 1994/51). The plan was approved by the National Security Council on November 30, 1994 and officially announced by Tansu Çiller. In December of that year the government contacted the Social Development Fund of the European Council for support. Çiller requested $278 million for the implementation of the project, but Turkey was soon compelled to abandon its application. The main reasons for the failure to raise funds—according to a Dutch diplomat, an anonymous source at the Dutch Ministry of Foreign Affairs, interviewed September 2, 2002—was the idea that Turkey was shifting the costs of village depopulation and destruction onto Europe. The scheme was considered the tail-end of a policy of forced migration, devised to keep the displaced permanently away from their former villages. Although these two schemes were not implemented, a few projects incorporating ideas enclosed in the toplu kondu and toplu çiftlik projects have been realized—not, however, in the form of urban entities established in a rural environment (which is also a basic idea of the village-town project, discussed in Chapter 4), but as out-of-town developments near urban
84
chapter two
centers. Examples are the construction of Be yüzevler, near the city of Diyarbakır; 75nci Yıl Toplum Merkezi—Yalimerez Mahallesi, near the city of Van; Doğankent; near the provincial center of Hakkari; and Kandolar Mahallesi, Afet Evleri (literally, Kandolar Quarter, Disaster Houses), also known as 80 Evler (80 Houses), near the a district town of Ovacik in the province of Tunceli. Be yüzevler was constructed for the resettlement of displaced persons from Lice, a district north of Diyarbakır where most of the rural settlements were evacuated. The Doğankent and 75nci Yıl Toplum Merkezi projects were established for people from Hakkarı, in particular from settlements near the border with Iraq—Doğankent was intended for inhabitants of the Uzungeçit area, and many of the 75nci Yıl Toplum Merkezi inhabitants were resettled from Uzundere (a settlement not officially recognized as evacuated: following its actual evacuation in 1995, Uzundere was ‘abolished’ in 1998 and thus evaded formal listing) (HRW 2005: 14). 75nci Yıl Toplum Merkezi has about 300 houses, of 60–70 m² on 500 m² plots. Ovacik has prefab housing, for people from evacuated and destroyed rural settlements in the proximity of the district town itself. Sources of income in these settlements are scarce. Most men remain unemployed and continue either to seek work as day laborers (e.g. most 75nci Yıl Toplum Merkezi men go into Van) or to roam the streets with barrows selling vegetables. Women make a small income knotting carpets, or are occupied in subsistence farming. With such poor prospects for daily survival in the settlement schemes, many people have left, or did not even move there in the first place (e.g. Uzungeçit evacuees did not settle in the Doğankent settlement scheme, but just headed straight for Van). Such projects were not implemented on a large scale, for two major reasons. First, the plans for scheming resettlement came at the tail-end of displacement policies, only entering the political agenda after the greater part of the mass displacement had already taken place. Second, resettlement schemes are physical symbols that reveal the magnitude and gravity of internal displacement (a matter successive Turkish governments have been at pains to avoid), while resettlement tracks, on the other hand, render the displaced largely invisible.
The evacuation of villages was organized in the form of what I term rural-to-urban resettlement tracks, defined as (multiple) routes from rural to urban settlement entities along which people were forced to move (without support or assistance from authorities). Armed forces evacuated and destroyed villages, mapping out the direction of movement, which was basically from the countryside to cities. In the cities, the displaced were left to their own devices. Most of them found a place to stay and shelter, mainly through chain-migration and self-help. The chain-migration mechanism implied that the evacuated selected urban centers which had already been established as settlement destinations by
soldiers and settlements
85
their relatives or hem ehri (people from the same place—village, district or region—of origin). The self-help implied that they would re-establish themselves in the urban entities through informal support networks, and without state assistance (or compensation for losses incurred during the eviction). The extent of the actual change in the rural-urban environment of the war may be illustrated by the data presented by Matthew Kocher (2004), using population counts from to show the change in ethnic geography for the period 1990–1997, when the majority of evacuations took place. The war-affected region is determined using two methods: first, the area falling under the State of Emergency, OHAL (Olağanüstü Hal ), legislation (thirteen provinces, at its peak); and second, the Kurdistan area, defined as those provinces with over 10 percent native Kurdish speakers according to the Turkish census of 1965 (in which the number of native Kurdish speakers was seriously undercounted)— twenty-one provinces.38 In gross terms, shown in Table 2.4, Kocher’s data reveal the rural population in the OHAL and the Kurdistan areas to have dropped by 11.9 and 10.3 percent respectively between 1990 and 1997, as against a general population growth in these areas for the same period of 14.0 and 12.8 percent. The urban populations increased drastically, by almost 45 and 40 percent. In Non-OHAL and in non-Kurdistan Turkey, the comparable figures show a decline in rural population by just 4.0 and 3.4 percent and an increase in the urban population by a little less than 21 and 20 percent. Defining the war-affected region by the two measures, OHAL and Kurdistan, does not show significant differences (and likewise for the non-OHAL compared with non-Kurdistan areas), whereas the differences between the war-affected area (however defined) and the rest of Turkey are obviously huge.39
38 The OHAL state of emergency area was created in 1983 (and the office of the Regional Governor of the region in 1987), maintaining military control in the Southeast after executive power was otherwise transferred back to the civilian authorities following the 1980 coup. Civilian rule was later returned to the area, albeit slowly and piecemeal, until the final lifting of OHAL law from the remaining four provinces and termination of the region in 2002. The OHAL organization was presented by the Turkish state as an administrative disaster management rather than counter-terrorist measure, established to deal with natural disasters and social unrest (like the US FEMA). A reestablishment of OHAL law continues to be mooted by politicians and the military (e.g. Korkut 2006). 39 Although not officially reported, village evacuations also occurred outside of the OHAL area (HRW 2005: 16). This would seem largely to explain the Kurdistan nonOHAL figures, which approach those of the Kurdistan and OHAL areas, but do not
86
chapter two
Table 2.4 Population growth in OHAL and Kurdistan Provinces in Turkey
(Source: Kocher 2004) Provinces
% ∆ Population % ∆ Rural % ∆ Urban Population 1990–1997 Population 1990–1997 1990–1997
OHAL (13) Non-OHAL (67) Kurdistan (21) Non-Kurdistan (59) Kurdistan Non-OHAL (8)
14.0 11.0 12.8 11.0 11.4
–11.9 –4.0 –10.3 –3.4 –8.3
44.5 20.9 39.8 19.9 34.3
The same dramatic changes in the social geography of the OHAL and Kurdistan areas for the 1990–1997 period (as compared to the rest of Turkey), are revealed when the population changes are presented in relative (rural-to-urban) terms (Table 2.5). In 1990, the majority (approaching 55%) of the Kurdish population lived in rural areas, but by 1997 this proportion had dropped to under half (less than 45%). Table 2.5 Proportion of the population living in rural areas
(Source: Kocher 2004) Provinces OHAL (13) NON-OHAL (67) Kurdistan (21) Non-Kurdistan (59) Kurdistan NON-OHAL (8)
% Rural 1990
% Rural 1997
% ∆ Rural 1990–1997
54.2 39.6 53.9 38.2 53.7
41.9 34.2 42.9 33.2 44.1
–14.0 –5.4 –11.0 –5.0 –9.5
The decline in the rural population) is serious (14% and 11% respectively as a proportion of the total populations of the OHAL and Kurdistan areas), and much greater than the trend for the rest of Turkey (around 5%). Clearly, however, many people continued to live in the countryside, which may lead to a questioning of the real extent of the rural ‘depopulation’—doubts assuaged when the regional distribution of the depopulation is taken into consideration. This is best illustrated by looking at the district rather than provincial statistics (Table 2.6).
reach the same extent. The Kurdistan non-OHAL area, that is, specifies a partially war-affected region.
soldiers and settlements
87
Table 2.6 Cumulative population growth in Diyarbakır districts
(Source: Kocher 2004) District (listed by ascending order of population) Kocaköy Çüngü Hazro Lice Eğil Hanı Kulp Dicle Çermik Çınar Ergani Silvan Bismil Province Center Province Total
% ∆ Rural Population 1990–1997
District Center 1990–1997 Cum. ∆ Rural Population
% ∆ Urban Population 1990–1997
District Center 1990–1997 Cum. ∆ Urban Population
–23.0 –20.4 –36.5 –73.5 –1.0 4.0 –56.4 –19.6 –1.7 –0.4 –4.8 –12.1 –15.1 48.2 –9.0
–2,567 –5,248 –11,058 –37,101 –38,729 –38,050 –62,323 –68,322 –68,885 –69,048 –71,022 –77,105 –86,027 –43,737 –43,737
19.3 5.2 –25.5 –20.2 –7.2 5.1 46.6 140.4 128.1 56.3 28.2 22.8 154.6 34.2 38.6
819 1,022 –1,029 –3,378 –3,722 –3,194 291 7,891 10,009 15,684 26,237 39,894 101,469 231,965 231,965
When changes in the rural and urban population disaggregated to district level (aggregated population statistics for provinces in the Kurdistan region are listed in Table 2.7), the statistics show that while in Diyarbakır province as a whole the rural population declined by some nine percent over the period in question, in about half the districts the reduction was at least double that, and as high as 73.5 percent in one district (Lice). Rural population numbers dropped dramatically in the districts of Lice, Kulp, Hazro, Kocaköy, Çüngü , and Dicle, all of which had relatively high numbers of mountain villages. In Hazro and Lice the rural de-population coincided with an urban de-population. In Kulp on the other hand, the rural de-population coincided with urban population growth. According to information collected by oral interviews with people in Kulp, the growth of the urban population there was caused—at least in part—by the withdrawal of village-guards to the district center from its outlying villages (e.g. see case study of İslamköy, Chapter 6).
437,093 344,121 249,074 330,115 1,096,447 498,225 299,251 1,010,396 172,479 704,359 558,275 243,435 133,584 262,006 1,001,455 637,433
Ağrı Batman Bingöl Bitlis Diyarbakır Elazığ Erzincan Gaziantep Hakkari Malatya Mardin Siirt Tunceli ırnak Urfa Van
158,758 194,664 86,648 144,029 595,440 274,045 144,144 738,245 71,522 369,243 249,032 110,221 50,799 125,264 551,614 262,562
278,335 149,457 162,426 186,086 501,007 224,180 155,107 272,151 100,957 335,116 309,243 133,214 82,785 136,742 449,841 374,871
528,744 456,734 253,739 388,678 1,362,708 569,616 316,841 1,285,249 236,581 853,658 705,098 263,676 93,584 353,197 1,443,422 877,524
Total 252,309 304,166 123,470 219,511 817,692 364,274 172,206 1,009,126 139,455 499,713 391,249 153,522 54,476 211,328 842,129 446,976
2000 City
Village
1990 City 276,435 152,568 130,269 169,167 545,016 205,342 144,635 276,123 97,126 353,945 313,849 110,154 39,108 141,869 601,293 430,548
Village 19.03 28.30 1.86 16.33 21.73 13.39 5.71 24.05 31.59 19.22 23.34 7.98 –35.58 29.86 36.55 31.96
Total
46.31 44.62 35.40 42.13 31.71 28.45 17.78 31.25 66.76 30.25 45.16 33.13 6.99 52.28 42.30 53.19
–0.68 2.06 –22.06 –9.53 8.42 –8.77 –6.99 1.45 –3.87 5.46 1.48 –19.00 –74.97 3.68 29.01 13.84
Annual Growth Rate % City Village
40
The name of the State Institute of Statistics, DİE (Devlet İstatistik Enstitüsü) has been changed since these statistics were issued, to the Turkish Statistical Institute (Turkstat), TÜİK (Türkiye İstatistik Kurumu).
Total
Province
Census Year
Census Year
(Source: State Institute of Statistics: DIE 1997)40
Table 2.7 City and village population and annual growth rate of population in the Kurdistan region by province
88 chapter two
soldiers and settlements
89
The net effect of the evacuation of rural settlements was an increased urbanization of the Southeast. In Diyarbakır, the urban population increased by 38.6 percent. And although the total population of the Southeast as a whole did not decline, the rate of increase did. Between 1985 and 1990, the annual population growth in the region was 3.49 percent, which fell to 2.43 in the period 1990–1997.41 Disaggregated to the level of provinces, it is again only the rate of population growth which declines, except for Tunceli where there is an absolute reduction of 35.6 percent for the province as a whole and 75 percent for the rural area (Table 2.8). The dramatic effect of the evacuations on the human geography of the region may also be illustrated by data on economic losses, although it must be emphasized that reliable statistics are lacking and numbers are necessarily estimations. According to a rough and non-specified figure provided by David McDowall (2000: 440), agricultural losses for the region amounted to as much as $350 million in the year 1994. The president of the Society of Agricultural Engineers (Ziraat Mühendi ler Odasi) in Diyarbakır valued the loss of live-stock alone in the decade 1990–1999 at $2.3 billion—in terms of meat production/consumption, Turkey went from a net exporting to a net importing country during this period (Tutkun 2002: 55). The sudden drop in animal husbandry and life-stock production was not only caused by the village evacuations, but also by a ban on the movement of animals to the summer grounds ( yayla yasağı) which seriously affected production capacity.42 Conclusions The PKK developed within a student and urban environment (the founding members of the party met in Ankara as housemates, in university classes, and left-wing student and youth organizations) but its strategy was mainly based on building up forces in the countryside under the protection of a guerrilla army. In contrast to ‘revolutionaries’, who aim to take over the state in a moment of great change and fight in the center of power in order to conquer it (an institutional strategy), the PKK used an insurgency strategy, a spatial strategy 41 In 1985 the population of Southeast Anatolia was 4,346,947; in 1990, 5,159,464; and in 1997, 6,128,973 (DİE 1997). 42 The yayla yasağı are mountainside villages/hamlets and grazing areas, high up and away from the heat, but also, of course, in prime PKK territory.
Total
City
Village
Total
City
2000 Village
General Total 56.473.035 33.656.275 22.816.760 67.803.927 44.006.274 23.797.653 Ağn 437.093 158.758 278.335 528.744 252.309 276.435 Batman 344.121 194.664 149.457 456.734 304. 166 152.568 Bingöl 249.074 86.648 162.426 253.739 123.470 130.269 Bitlis 330.115 144.029 186.086 388.678 219.511 169.167 Diyarbakır 1.096.447 595.440 501.007 1.362.708 817.692 545.016 Elaziğ (2) 498.225 274.045 224.180 569.616 364.274 205.342 Erzincan 299.251 144.144 155.107 316.841 172.206 144.635 Gaziantep 1.010.396 738.245 272.151 1.285.249 1.009.126 276.123 Hakkari 172.479 71.522 100.957 236.581 139.455 97.126 Malatya 704.359 369.243 335,116 853.658 499.713 353.945 Mardin 558.275 249.032 309.243 705.098 391.249 313.849 Slirt 243.435 110.221 133.214 263.676 153.522 110.154 Tunceli 133.584 50.799 82.785 93.584 54.476 39.108 ırnak 262.006 125.264 136.742 353.197 211.328 141.869 Urfa 1.001.455 551.614 449.841 1.443.422 842.129 601.293 Van 637.433 262.562 374.871 877.524 446.976 430.548
Province
1990
Census year
(Source: State Institute of Statistics: DIE 1997)
18,28 19,03 28,30 1,86 16,33 21,73 13,39 5,71 24,05 31,59 19,22 23,34 7,98 –35,58 29,86 36,55 31,96
Total
26,81 46,31 44,62 35,40 42,13 31,71 28,45 17,78 31,25 66,76 30,25 45,16 33,13 6,99 52,28 42,30 53,19
City
%
4,21 –0.68 2,06 –22,06 –9,53 8,42 –8,77 –6,99 1,45 –3,87 5,46 1,48 –19,00 –74,97 3,68 29,01 13,84
Village
Annual growth rate population
Table 2.8 City and village population and annual growth rate of population in the Southeast by province
90 chapter two
soldiers and settlements
91
involving the gradual development of a counter-power and a counterstate, with a building up of forces where the State is weakest, i.e. the peripheral countryside. Initially, the Turkish armed forces contributed to the fulfillment of the PKK strategy, in particular as a consequence of the decision to defend larger villages and towns. By ignoring the smaller rural settlements, the Turkish armed forces created the space for the PKK to establish a guerrilla network throughout the Southeast. However, its dependence on this space is also a guerilla’s weak point, and spatial deprivation the Achilles heel. The ‘field domination doctrine’ announced by the General Staff in 1991 and systematically implemented from 1993 onwards changed the situation dramatically. The objective of the new doctrine was the destruction of the PKK environment, both by contraction (resettlement of the population) and penetration (deployment of special forces, applying the principles of a war of movement, and penetrating the spaces of the PKK, as well as drafting the civilian populations in PKK areas into the village guard system). At a tactical level, the resettlement and drafting policies both denied the guerrilla food, shelter, intelligence and recruits, and created kill-zones in the countryside. At a strategic level, the army engineered a new settlement pattern by accelerating ruralto-urban migration (under duress, by threat and intimidation and by burning houses and destroying villages), thereby forcing the guerrilla to choose between retreat or engagement in a confrontation with the state in urban entities (a tough environment for insurgents, but favorable for the state). The main conclusion of this chapter is that the evacuation of thousands of small rural settlements in the countryside was not collateral damage or reprisal, but a constituent part of the field domination counter-insurgency strategy. The resettlement of the population was not openly declared, since this practice was outlawed in 1977 under the ratification of the 2nd Protocol Additional to the Geneva Conventions. However, high ranking military personnel openly hinted at a large scale evacuation of rural settlements and the concentration of the population in a few large settlements. A striking feature of the resettlement effected by the Turkish armed forces was the absence of any policy of resettlement schemes. The resettlement of villagers took the form of tracks, defined as pathways of forced migration from rural to urban entities (without any support or assistance from the authorities). This made resettlement in Turkey in the 1990s not only inherently cynical, but also externally invisible. The displaced Kurds dispersed and, in spatial terms, vanished.
CHAPTER THREE
SETTLING THE SOUTHEAST The Rise and Fall of a Rehabilitation Doctrine Apart from the social and economic problems, the event of evacuated villages in East and Southeast Anatolia has created new opportunities and dynamics for the formation of new standards that can accomplish a new rural settlement pattern; for the transition from dispersed and unsuitable settlement units towards settlements units of sustainable size and potentials. Oğuz Oyan (et al.) The millions of displaced people (. . .) are nothing but refugees of an unacknowledged war. And we are condoning it by looking away. Why? Because we’re told that it’s being done for the sake of the greater common good. That it’s being done in the name of progress, in the name of national interest. Therefore gladly, unquestioningly, almost gratefully, we believe what we’re told. We believe that it benefits us to believe. Arundhati Roy1
Introduction The military authorities, and not the civil authorities, controlled the evacuated territory in southeast Turkey in the 1990s. Things had gone so far that on several occasions top-level politicians were refused access to areas where resettlement operations had been carried out. The military prevented Prime Minister Tansu Çiller from visiting Lice in October 1993, and Deputy Prime Minister Murat Karayalçin from visiting Hozat in October 1994, while in 1997, Hüseyin Yıldız, a member of the Parliamentary Commission on Migration, was prevented by the Governor of Mardın and the chief of police from going to Çınarönü. Nevertheless, civil authorities did gradually become involved in the issue of resettlement. In 1997, the Commission on Internal Migration, headed by Ha im Ha imi, Diyarbakir MP for the Welfare Party, RP (Refah Partisi), put the nightmare of the army’s resettlement policy squarely on record (HRW 2002). The Commission was initially established to 1
Oyan et al. 2001: p7 and Roy 1999.
94
chapter three
deal with the effects of migration to the cities, but extended its terms of reference, studied the evacuation of villages, and advised an unconditional and voluntary village return (Meclis Ara tırma Komisyonu 1997).2 The issue of village return had entered the political agenda already in 1995, when a coalition government of the True Path Party, DYP (Doğru Yol Partisi) and the Republican People’s Party, CHP (Cumhuriyet Halk Partisi) proposed a return-to-village program in the context of the Southeast Restoration Project, GAP (Güneydoğu Onarım Projesi). Not much information is available about this program, although it was reportedly a blueprint for the organization of a gradual return to those evacuated villages where security could be provided. Apparently, in order to mitigate the economic burden in reconstructed villages, the program was to provide funds to support traditional agrarian occupations, such as bee keeping and animal husbandry for men, and carpet weaving for women. The proposed program was not exactly trail-blazing, but from a political perspective it was significant, most especially in that it involved the idea that villages which had been evacuated and destroyed had to be eligible for return (an idea opposed by the military and governors in the region, who claimed that a re-habitation of evacuated villages would create a security risk). In 1997 the leader of the Motherland Party, ANAP (Anavatan Partisi) and the then prime minister, Mesut Yılmaz, also announced that his government would support the return of evacuated populations to their villages. Again, return was made conditional on the ability of the army to provide security, but the implicit, and fundamental, purport was that the evacuation of 2 The establishment of the Commission should be seen partly as the result of international pressure, but also in the context of domestic politics, set against the ongoing tensions between the military and civic authorities in general and the government of the day in particular. Heading the government as the major coalition partner when the Commission was set up, the Welfare Party had a significant conservative Kurdish franchise. It was also the first Islamic party to form a government in the history of the Republic of Turkey, and as such, was in constant conflict with the army during its short spell in office after winning the 1996 election. The ideological clash between the Kemalist armed forces and the Islamicist politicians was negotiated through political maneuvers and meetings of the military-civil executive body, the MGK. Finally, the Welfare Party was forced out of power in the ‘post-modern’ coup of 1997, its leader, (ex-)Prime Minister Necmettin Erbakan, served a life ban from politics, and the party formally outlawed in 1998—after which it re-emerged as the Virtue Party (Fazilet Partisi), but having already lost the leadership role in Islamic politics in Turkey to Tayip Erdoğan’s AKP.
settling the southeast
95
the countryside was an anomaly, and return inescapable. And again, nothing happened. This chapter will discuss the most recent plans for the development of a framework in which return was supposed to take place, namely the village-town (koy-kent) and the center-village (merkez-koy) models. These plans will be discussed from an institutional perspective, as attempts to reorganize rural space in the conflict areas. The argument is that different state institutions have different spatial strategies, i.e. they think differently about the organization of rural space. Institutional thinking We have defined the state as a set of interconnected institutions concerned with the administration and control of a geographically determined population and backed by the capacity to use force to implement its policies (Chapter 1). Like any other type of institution, state institutions produce and maintain ideas, an issue elaborated upon by Mary Douglas in her book, ‘How Institutions Think’ (Douglas 1986). Of course, as Douglas freely admits, institutions do not really think— rather, it is people in the institution who do the thinking (and develop and maintain ideas), and it would, therefore, be more correct to talk about institutional thinking (of social actors) (see Douglas 1989). Drawing primarily upon the work of Emile Durkheim and Ludwik Fleck, Douglas explains the social origins of individual thought (i.e. that it is constituted within collective practices, referred to as institutions). Institutions do not only produce and maintain ideas, but in pursuit of these ideas also enable people to act collectively (Becker 1999). In other words, institutions lead people to work in particular ways in order to materialize ideas, conceptions that are themselves importantly institutional in origin. The idea of ‘working-concepts’ synthesizes the notions of (institutional) thinking and doing. Such working-concepts are nothing more than practical applications of institutional thinking, guiding people in their day-to-day actions and the collective nature of their endeavors. At its broadest, the institutional thinking of the Turkish Armed Forces in respect of the PKK was to view it as an enemy to be defeated, by whatever means, which came to be crystallized as the doctrine of field domination. The working concept related to this doctrine to guide
96
chapter three
actions on the ground was (and indeed, still is) that of draft or destroy, which involved both a low-level militarization (turning villagers into paramilitaries) and evacuation of the countryside (turning villagers into displaced persons). This chapter will argue that the institutional thinking of civil institutions may be grouped under the general heading of a rehabilitation doctrine. This doctrine was linked to the working concepts of village-town and center-village, which were based on the idea that rural space has to be reorganized by means of horizontal and vertical integration (of places and people into the spatial, economic, social, and cultural landscapes of the nation-state). It will be argued that the materialization of ideas in the working concept of field domination on the one hand and village-town and center-village on the other, and the strain between these two, very different concepts, resulted in a failure of the rehabilitation doctrine. Note on sources A number of methods were used for the collection of data for the study of the rehabilitation doctrine, and specifically, the village-town and center-village models. These included formal and informal interviews, field trips, and the study of reports and secondary literature. Unpublished documents form the basis of the study of, in particular, the center-village model and the East and Southeast Anatolia Return to Village and Rehabilitation Project Sub-Regional Development Plan (Doğu ve Güneydoğu Anadolu Bölgesi Köye Dönü ve Rehabilitasyon Projesi Alt Bölge Geli me Planı)—henceforth the Village Return and Rehabilitation Development Plan. This plan is certainly the most important such document to be considered here. The research for the Village Return and Rehabilitation Development Plan was coordinated by GAP and guided by a steering committee composed of the Minister of Internal Affairs, the (twelve) governors of the war affected provinces, the State Planning Organization, DPT (Devlet Planlama Te kilatı) and the General Directorate of Rural Services, KHGM (Köy Hizmetleri Genel Müdürlüğü). Research was supervised by and written under the auspices of Prof. Dr. Oğuz Oyan and Prof. Dr. Melih Ersoy. At the time of the study, Oğuz Oyan was president of the Turkish Social Sciences Association, TSBD (Türk Sosyal Bilimler Derneği) and vice-president of the Republican People’s Party, CHP (he was
settling the southeast
97
elected to Parliament as a CHP MP in 2002); Prof. Dr. Melih Ersoy was professor at the Department of Urban and Regional Planning, Faculty of Architecture, Middle East Technical University, ODTÜ (Orta Doğu Teknik Üniversitesi), in Ankara. Under the supervision of Oyan and Ersoy interviews were carried out with governors, deputy-governors, district officers and mayors, and ‘focus group’ interviews organized. Focus group interviews are based on facilitated discussions with a group of individuals selected because they are believed to be representative of some category—in this case, displaced persons. The method is centrally concerned with understanding attitudes rather than measuring them. It approaches attitudes and priorities tangentially by allowing respondents to talk freely (Greenbauw 1998). Focus group interviews were carried out in 297 villages in eleven of the war-affected provinces (Batman, Bingöl, Diyarbakır, Elazığ, Hakkari, Mardin, Mu , Siirt, ırnak, Tunceli, and Van), involving 1,097 people (most of them displaced villagers). The villages were selected from three different lists provided by national, regional and provincial authorities (the Government, GAP and governors), nominated as potential locations either for concentrated settlements or for the provision of center functions (serving the surrounding, evecuated settlements). Essentially, the Village Return and Rehabilitation Development Plan is constituted by twelve volumes, one per (war-affected) province (the eleven provinces where focus group interviews were carried out, plus Bitlis). Each volume, of 100 to120 pages, contains four parts. The first part, the Definition and Scope of the Return to Village and Rehabilitation Sub-Region Plan (Köye Dönü ve Rehabilitasyon Alt Bölge Planının Tanımı ve Kapsamı), is repeated in each volume and sets out the conceptual framework. The second part, Planning Organization and Focus Group Activities (Planlama Çalı masının Yöntemi ve Odak Grup Çalı maları), gives the results of the focus group interviews, including quantitative information on pre- and post-migration work and income, and qualitative information in the form of opinions concerning the return-to-village process, the support expected from the authorities, and ideas about a future, post-return reality. The third part, ‘A Sub-Regional Development Plan’ (Alt Bölge Geli me Planı Yöntemi ve Raporu), is the most extensive (covering almost half of each report), and comprises a feasibility study which assesses the local socio-economic, agricultural, geological, and climatologic variables. The fourth and final part is an investment action plan. Parallel to the twelve provincial volumes, a Summary for
98
chapter three
Administrators (Yönetici Özeti) was prepared for each province, serving, in effect, as proposals for pilot projects (Oyan et al. 2001b, c, d, e, f, g, h, i, j, k, l, m). These pilots include an assessment of development potentials, an action plan and a budget, mainly for road construction, the supply of drinking water and electricity, and the construction of boarding schools for children, and Turkish language and handicrafts education centers for Kurdish women. There are other related, and unpublished, documents, such as the Return-to-Village and Rehabilitation Project (Köye Dönü ve Rehabilitasyon Projesi) (GAP 2000) and the Return-to-Village and Rehabilitation Project, Return-to-Village Urgent Action Plan (Köye Dönü ve Rehabilitasyon Projesi, Köye Dönü için acil uygulama Programı) (Office of the Prime Minister n.d.). In the document produced by GAP, the concept of return and rehabilitation is discussed abstractly, as a rural development strategy supposedly contributing to a balanced distribution of the population and the establishment of center-villages. The document prepared by the Prime Minister’s Office is merely an incomplete list of villages where return had allegedly taken place, but lacking any details. Discussion here of the projected return to and rehabilitation of the evacuated rural settlements therefore relies on the Village Return and Rehabilitation Development Plan. The village-town model General backgrounds The village-town model entered the political agenda with the rise to power of the senior politician Bülent Ecevit in 1999, and was dropped again after his fall in 2003. For the greater part of his long political life (he was first elected to Parliament in 1957), Ecevit had thought that the village-town model could change the future of Turkey. The model comprised a reorganization of rural space by means of a simultaneous horizontal and vertical integration of rural settlements. The vertical (administrative) integration of rural settlements aimed to establish the state as a stable center. The horizontal integration was supposed to turn self-contained villages, under the influence of superstition and conservatism, into an integrated part of the nation-state, loyal to the principles of Atatürk and engendering close cooperation with the state (see Chapter 4 for a further discussion of the model). Simply, it was
settling the southeast
99
considered that spatial integration would provide a common basis for political and social life. Bülent Ecevit had gained power three times in the government carousel of the 1970s—for about ten months in 1974, one month in 1977 and eleven months in 1978—and preparations to implement the village-town model had finally materialized in 1978 with two pilot projects in Van and Urfa (both provinces in the Kurdistan region in Turkey). The projects were terminated with the fall of his government in 1979, however, and one year later, after the 1980 military coup, it seemed this would be the fate of Ecevit’s political career also, as he was banned from politics (along with the prominent leaders of other political parties). Yet, within a few years, he had returned, founding a new party, the Democratic Left Party, DSP (Demokrat Sol Partisi). Initially, the party was headed by his wife, Rah an, but eventually the military backed off and the ‘generation of the 1970s’ returned into politics, among them Ecevit The DSP, tightly organized by husband and wife, played a rather insignificant role in politics throughout the 1990s. An expeditious, albeit short-lived, comeback started when Ecevit was asked to form a caretaker government, which took office on January 11, 1999. Later that year, after the May 28 elections, the DSP became the biggest party in Parliament, mainly as a result of voter aversion to the parties then in power and what may be called ‘the Öcalanbonus’—on February 16, 1999 prime minister Bulent Ecevit startled the country with the news that the Kurdish rebel leader Abdullah Öcalan had been captured in Kenya the day before in a joint operation between the U.S. Central Intelligence Agency, the CIA, and the Turkish National Intelligence Agency, MİT (Milli İstihbarat Te kilatı) and flown to Turkey.3 Following his election victory, Bülent Ecevit formed a coalition government—his leftist-nationalist DSP being joined by the right-wing nationalist National Action Party, MHP (Milliyetçi Hareket Partisi), and the conservative-liberal Motherland Party, ANAP (Anavatan Partisi)—and on June 4, 1999, presented his government’s program. In this presentation, he referred the government’s intention to establish village-towns (without using the term as such), and associated this with a return to the villages in the Southeast (Ecevit informed Parliament that his government would establish coordination and cooperation between
3 Allegedly, Turkish special service agents captured the PKK leader after his cell phone call locations had been tracked by U.S., British, and Israeli intelligence agents.
100
chapter three
rural settlements located in close proximity and accelerate the process of village return).4 The Çavdar-Mesudiye and Dorutay-Özalp Projects In a letter to the Prime Minister dated January 5, 2000, the Hürriyet journalist Oktay Ek i, a confidant of Ecevit, proposed his hometown of Mesudiye as a suitable site to accommodate a village-town pilot project. Mesudiye is a rather inward, high altitude and not easily accessible town in the north of Turkey. Located in the province of Ordu, the district of Mesudiye is administratively dependent provincial capital, the town of Ordu on the Black Sea coast (although the inhabitants of Mesudiye consider themselves ‘inner Anatolians’). A single road connects Mesudiye with Ordu, meandering along and high above ravines as it passes through the mountains. The 120-kilometer distance can be completed in three hours by an experienced driver. On a development index of district towns in Turkey, Mesudiye occupies place 632 out of 858 (Anonymous 2000: 4). Measured in terms of per capita income, Mesudiye was placed 575 out of Turkey’s 918 districts (Oyan et al. 2001a: 91–5).5 This was explained by Mesudiye’s rural character—in Turkey the rural population was about 40 percent, but in Mesudiye it was about 80 percent (Eksi 2001: 4). The mountains accommodated militants of the Liberation Army of Turkish Workers and Peasants, TİKKO (Türkiye İ çi Köylü Kurtulu Ordusu), the armed wing of one of Turkey’s illegal communist parties. Bülent Ecevit gave a positive response to his confidant on June 12, 2000, and a feasibility study was prepared just two weeks later. This feasibility study, into the construction of a village-town in Mesudiye, grouped the rural settlements in the district into nine clusters, which was later reduced to eight.6 The Çavdar cluster of nine rural settlements 4 The village cooperation and coordination plan announced in the coalition program was intended to make urban opportunities and possibilities available at village level. It was also intended to speed up return-to-village by those who had been forced to leave their homes as a consequence of terrorism: ‘Birbirine yakın köyler arasında i birliği ve e güdüm sağlanarak, altyapı hizmetlerinin ve sosyal hizmetlerin tüm köylülere daha dü ük maliyetle ula ması gerçekle ecek ve böylece köylüler kentlerdeki olanaklara yerlerinden olmaksızın kavu abileceklerdir. Terör nedeniyle yerlerini terketmek zorunda kalmı yurtta larımızın köye dönü olanakları da hızlandırılacaktır.’ 5 In 1996, average per-capita income was $3,021 in Turkey as a whole, $1,543 in Ordu province and $897 in the Mesudiye district. 6 These were: Arıkmusa-Güneyce (ten rural settlements), Bayırköy-Mahmudiye
settling the southeast
101
was chosen as the site of the first village-town, partly because it was said to be the least developed of the eight clusters, but also because its inhabitants tended to religious conservatism. The planned transition of the sub-district’s economy from agrarian to agro-industrial, and concomitant increase of prosperity, it was anticipated, would lead these people to disassociate themselves from their religious conservatism (in a manner similar to that in which a transformation of the Southeast was supposed to disengage people from their ‘tribal’—read Kurdish—identity). The population of the Çavdar cluster was put at 3,023, although it was acknowledged that the actual population was rather elastic (the summer population being double that of the winter—many people in the area had left to İstanbul and Germany, using the district as their summer resort (TKV 2000: 9, 20).7 The village-town plan did not come cheap. Construction of roads was a key component of the Çavdar project and accounted for the bulk of the costs, which were estimated at $10 million (the amount assumed as the average cost of a village-town project).8 In earlier times, a rude calculation had led to the estimation that the clustering of all rural settlements in Turkey would require a total of some 5,000 village-towns at a cost of $50 billion (Bulguç 1976). Given the financial resources at the disposal of the Ecevit government, his expansive vision would remain an expensive dream without outside investment, so a loan was applied for from the World Bank. The World Bank gave a positive response and released credit to the tune of $300 million on November 28, 2002.9 Half of the loan was to be expended on improving the rural infrastructure—specifically, the roads and highways—with the other half to be invested in the improvement of social services, such as development of the education sector and income generating activities (World Bank 2002: 1). At an average of $10 million per project, the World Bank credit would allow the (eleven rural settlements), Çavdar (nine rural settlements), Mesudiye (eight rural settlements), Topçam (seven rural settlements), Üçyol (six rural settlements), Ye ilçe (six rural settlements), and Yukarı Gökçe (eight rural settlements) (Anonymous 2000: 4; Basa and Eksi 2002: 4). 7 The 3,023 figure was made up from: Çavdar 2,168 inhabitants, Esatlı 130, Türkköyü 74, Kıslacık 71, Çardaklı 161, Göçbeyi 134, Dayılı 166 and Yuvalı 119. 8 (Anonymous) coordinator of village-town projects at the Office of the Prime Minister, 10.10.2002 (personal communication). 9 The Mesudiye pilot project had been ceremonially initiated on September 2, 2000; its ‘opening’ ceremony on October 27, 2001 was attended by Prime Minister Ecevit and the representative of the World Bank in Turkey, Ajay Chhiber.
102
chapter three
establishment of thirty village-towns, and with the loan being planned to be released in equal amounts over a period of three years, ten projects a year were envisaged, starting in 2003. This posed a serious problem for the government, since it said it had received 250 applications from those wishing to replicate the Çavdar-Mesudiye village-town experience. The World Bank released the first installment of the loan on November 28, 2002, but by that time early elections had already resulted in the fall of the Ecevit government—his party was decimated, failing to pass the ten percent threshold and deprived of parliamentary representation (even in Mesudiye, the DSP only received a few votes). One of the first decisions of the new government formed by the conservative-Islamic Party of Justice and Development, AKP (Adalet ve Kalkınma Partisi or AK Parti), which had won a landslide victory, was to end the village-town plan and to return the $300 million loan to the World Bank.10 The new government also left the model project in Mesudiye-Çavdar to its own devices, but in truth, the model project was already in decay shortly after its opening. Although road construction was a key component, road connections in the Mesudiye-Çavdar village-town remained poor and construction was abandoned after the ceremonial opening of the project. In Türkköy, one of the villages within the Mesudiye-Çavdar village-town, a brand-new school was built, but remained empty, as did the library, cultural center, and health clinic. The village-town sporting complex, where the small stadium had been used for the opening ceremony of the village-town, and which offered football, basketball, and volleyball facilities, also went unused and had weeds growing rank in the playing fields just a year after the opening ceremony.11 To stimulate economic activities, a Village-town Forest Products Factory (Köy-kent Orman Ürünleri Fabrikası) was opened in 2001, but by 2002 only two of the original fifteen workers were still employed there and in 2003 the factory was closed. There had been problems with maintenance of machinery from the start due to a lack of qualified personnel. Furthermore, the accessibility and transport problems meant that the factory was unable to find a market for its products.12
An ‘agricultural town’ project was announced, which did not aim at the development of a village group, as in the village-town project, but at the unification of several rural settlements into new villages, to be constructed in ‘packages’ of 100, 200, or 300 houses. A location in Batman was selected as a pilot, but no action followed. 11 Throughout a journey there (October 3–5, 2002), I did not see more then ten people—it looked like a ghost town. 12 The factory produced furniture, including school-desks, and at the time of my stay, 10
settling the southeast
103
The Office of the Prime Minister, in consultation with the World Bank, planned one other village-town project in addition to MesudiyeÇavdar, in Dorutay, a sub-district town in Van-Özalp. The province of Van is located in the east of Turkey, on the border with Iran, and is divided today into twelve districts (including Özalp) with a further ten sub-districts (including Dorutay) containing 576 villages and 401 hamlets. The population of the province is 772,132 (see Table 3.1). Table 3.1 Districts and number of sub-districts, villages and hamlets in Van
(Source: Van Varliliği 2006) District Van Central District Bahçesaray Ba kale Çaldıran Çatak Edremit Erci Geva Gürpinar Muradiye Özalp Saray Total
Sub-district Village 2 – 1 – 1 – 2 – 3 – 1 – 10
90 19 63 64 26 12 85 4 71 39 52 23 576
Hamlet
Population
34 43 61 27 90 1 39 30 81 11 20 11 401
288,794 17,415 49,676 48,155 21,034 15,696 145,229 29,629 37,797 44,287 54,143 20,277 772,132
The district of Özalp is located in the east of the province close to the border with Iran, with which it has a road and railway links. It received its district (ilçe) status in 1948, although the town is said to have been founded as an administrative center in 1869, under the name of Mahmudiye. Allegedly, it was renamed Kazımpa a in 1932, after Kazım Pa a (General Kazım), one of the heroes of the War of Independence and who became president of the Parliament in the early 1930s. After the Surname Law of 1934 (which introduced surnames to the republic, not an Ottoman tradition), Kazım Pa a chose the surname Özalp (meaning Real Hero), and this eventuated in the renaming of the town, in the workers had just finished picnic-sets with an inscription of the i li municipality of Istanbul. These picnic sets were not going to be transported to i li, the richest district of Turkey, however. The mayor of i li, Mustafa Sarıgül, at the time member of Bülent Ecevit’s party, the DSP, had placed an order to support the village-town factory and donated the sets to the Mesudiye municipality, which simply piled them up.
104
chapter three
1937. Although the places Mahmudiye and Kazımpa a were identical, Özalp is different. Together with the change of name from Kazımpa a to Özalp, the administrative center was moved inwards, away from the border, to a new location, the village Kargalı, which today is a neighborhood of Özalp (Mahmudiye/Kazımpa a are identical to the settlement now named Saray). The rationale behind the movement of the administrative center was a reorganization of space for reasons of security. It followed a land exchange and border adjustment agreed between Turkey and Iran. The exchange of territory between Iran and Turkey was intended to give both countries better access to the Ağri Mountains, the scene of a Kurdish uprising. Had it not been moved, the administrative center, Kazımpa a, would have become a bordertown, which was deemed a security risk (Özgen 2004: 5–7). Dorutay is the only settlement with the status of a sub-district in Özalp and was considered as the center of a village-town project in 1978 as well as in the period 1999–2002. A particular feature of the sub-district is the presence of two ‘Turkish’ villages, Dönerdere and Emek. The inhabitants of these two villages originate from the village Uzungöl in the Çaykara district in Trabzon province. Their settlement history dates back to 1961, when then Prime-Minister, İsmet İnönü, and Finance Minister, Ferit Melen, who originated from Van, visited the Çaykara district after it had been hit by flooding. The decision was taken to permanently evacuate the inundated area and resettle the population hundreds of kilometers to the east, in Van-Özalp, which had plenty of uncultivated and productive land. The Dönerdere village was established with the construction of 85 houses and Emek village with the construction of 15 houses, a total of 100 houses for the settlement of 400 people. Dönerdere and Emek were supposed to function as hearths, a concept involving the establishment of ethnic Turk settlers in ethnically non-Turk areas, intended to contribute to a modernization of agriculture and the spread of ‘Turkish’ culture ‘from within’ (ethnic Turk settlers were established in Diyarbakir province in the 1930s for the same reason, an issue discussed in Chapters 5 and 6). Turkey and World Bank agreed upon the implementation a village-town project in Dorutay in the Özalp district of the Van province, and preparations for the establishment of a village-town of Dorutay-Özalp were initiated in 2001, leading to road construction and the establishment of a new school and village-hall. However, the project was abandoned after the rise to power of the AKP, and with it the efforts to cluster the settlements as a village-town, integrating the participating settlements and people into the national economy, administration, and culture.
settling the southeast
105
Organization The implementation of the pilot in Mesudiye-Çavdar was organized by the Office of the Prime Minister, illustrating the importance attached to it by Prime Minister Bülent Ecevit, and its continuation was to be overseen by a Project Management Unit (PMU), to be established under the authority of the General Directorate for Rural Services, KHGM (Köy Hizmetleri Genel Müdürlüğü). The PMU, it was intended, would be headed by a director and staffed by a small number of project consultants, who would assist the PMU director in day-to-day management. The unit would make the budgetary estimates, monitor the implementation and progress of the project, provide quarterly implementation reports and activity plans and disseminate project results. The KHGM fell under authority of the Office of the Prime Minister, its remit and organization specified in Law 3202 passed on May 9, 1985, which made it responsible for the mapping, maintenance and development of road infrastructure, the maintenance and development of clean and sufficient water supplies and the construction of waste water facilities, and the installation of electricity. The Directorate was also made responsible for the execution of resettlement policies (mainly the preparatory infrastructure work). The director of the KHGM was also given the task of establishing cooperation and coordination between institutions in the public sector for the enforcement of national security policies and development plans. Historically, the KHGM office was the continuation of the General Directorate for Settlement (İskan Umum Müdürlüğü) and the General Directorate for Land and Settlement Works (Toprak ve İskan İ leri Genel Müdürlüğü), which was responsible for the implementation of the Settlement Act 2510 of 1934 (discussed in Chapter 6).13 Although the KHGM PMU was in charge of the village-town projects in Turkey as a whole, in the resettlement zones in the Southeast it was the military which was in control. As a general rule, the military did not support the implementation of village-town projects in areas containing
13 The KHGM, with its many provincial departments, experience in village works and resettlement, and responsibilities for the coordination of the village-town projects, was later abolished (on January 13, 2005), by the AKP government under Tayyip Erdoğan: the maintenance and development of infrastructure was transferred to the provincial authorities, while settlement issues were considered of national interest and transferred to the Ministry of Public Works and Settlement. In 2002, the period under current discussion, the KHGM was in the middle of negotiations with the World Bank regarding the establishment of a PMU and the implementation of village-town projects.
106
chapter three
or near to evacuated settlements, and there are strong indications that the military had their reservations against the village-town as a viable model for development of the Southeast in general. Anyway, the military simply rejected any spatial approach aiming at a reconstruction, even partial, of the evacuated settlements and a return of the civilian population. The military favored concentrated settlements. In fact, the Turkish Armed Forces had been actively involved in the construction of a compact settlement in the area (between the villages of Yalınca, Sıcaksu, Akdoğu and Kılıçtutan). The foundations of this project had been laid by the commander of the 2nd Army in 2000, but the project was aborted in the summer of 2001 after army units were withdrawn from the area (Arisoy 2002: 94). A former commander of the Artillery Battalion of 61st Internal Security Brigade in Van was plainly negative about the Özalp-Dorutay village-town project, concluding that the village-towns were expensive but ineffective. The village-town Dorutay had not initiated social-cultural change in the area, the officer remarked, and people were ‘adverse’ (‘tepkisel’) to the State, as shown by their majority vote in the 2002 elections for DEHAP, the main Kurdish party at the time (Arisoy 2002: 97–9). Indeed, when village-town project of Özalp-Dorutay was abandoned in 2002 after Bülent Ecevit fell from office and the village-town program discontinued, many villagers believed that the reason for the policy change was local, because they had given their votes to DEHAP.14
The Democratic People’s Party, DEHAP (Demokratik Halk Partisi) was a left-of-center party advocating (Kurdish) civil rights formed in 1997 following a serial opening and banning of Kurdish/civil rights parties through the 1990s: the People’s Labor Party, HEP (Halkin Emek Partisi), 1990–93; Democracy Party, DEP (Demokrasi Partisi), 1993–94; and then the People’s Democratic Party, HADEP (Halkin Demokrasi Partisi). This history of Kurdish/civil rights political parties was initiated in 1990 by a break of Kurdish MPs from the then Social Democrat Populist Party, SHP (Sosyal Demokrat Halkçi Parti)—which then emerged as the Social Democrat People’s Party, also SHP (Sosyaldemokrat Halk Partisi) and became junior partner to the Demirel-Çiler governments; the history was defined by the conflation of the Kurdish/civil rights parties with the PKK in the eyes of the Turkish establishment, which led to the (military instigated, constitutionally declared) bans on the HEP, DEP and later HADEP. In the 2002 election, DEHAP managed to poll over 6% of the national vote, but which was insufficient to take it past the 10% threshold for parliamentary representation. It did, however, continue its predecessors’ successes in achieving considerable political power at municipal level throughout the Southeast, and beyond. In 2005, DEHAP was finally merged into the Democratic Society Party, DTP (Demokratik Halk Partisi), founded by Leyla Zana after her release from prison along with three other former DEP MPs who been jailed for alleged treason and PKK membership a decade previously (Leyla Zana had dramatically proclaimed her Kurdish identity, in Kurdish, at her swearing in ceremony as an SHP MP in 1991). 14
settling the southeast
107
Although the Özalp-Dorutay village-town project was not in a waraffected area, the civil authorities did envisage village-town projects in evacuated areas, and in this they not only faced opposition from the military: the World Bank opposed the implementation of village-town projects as village return projects, and consequently wanted exclusion of the evacuated areas from the area of implementation of the project to which it was lending its name and money. The vice-president of the KHGM stated that ‘international organizations’ (he referred to the London-based Kurdish Human Rights Project in particular) were pressurizing the World Bank to withhold support for implementation of the village-town projects insofar as they were to be used as village return projects, even though, the vice-president emphasized, the villagetowns did not aim at a spatial concentration of population, but local development. Pressure from international non-governmental organizations was confirmed by the World Bank.15 Not only the KHRP, but also organizations such as Human Rights Watch urged the World Bank to withhold support for village-town projects in the Southeast (on the basis of an inconsistency with the UN Guiding Principles on Internal Displacement and the rights of internally displaced communities) (HRW 2002: 8; KHRP 2002: 106). The World Bank had assessed the village-town model as a rural development tool (World Bank project information file).16 The villagetowns were supposed to improve infrastructure services, social services, and support income-generating activities, thereby contributing towards rural development in Turkey. The ‘working concept’ that guided the work on the ground was mainly construction (of roads and education/ administrative facilities). In the Southeast however, the main work to be done was the complete construction of new rural settlements. Ecevit’s government had thought that the village-town model could create the conditions for a re-population of evacuated areas in the Southeast, forging national and territorial integration in a region badly affected, sometimes devastated, by insurgency and counter-insurgency: there were serious doubts whether such a program would be in accordance with UN standards. Furthermore, the domain of conflict-induced resettlement falls under the province of the United Nation High Commissioner for Refugees UNHCR, which had no presence in Turkey. The World Bank
Sudipto Sarkar, August 13, 2002 (personal communication). This project information was originally available on the World Bank Internet site, at www.worldbank.org, but it has since been removed. 15 16
108
chapter three
covers only development-induced resettlement (Cernea and McDowell 2000; Muggah 2003a; Voutira and Harrell-Bond 2000: 57). The absence of the UNHCR and concern with violating the UN Guiding Principles on Internal Displacement thus led to the decision by the World Bank not to become involved in projects the areas where the standard (World Bank) working practice of infrastructure construction was not valid, i.e. in the evacuated zones of the war-affected areas. The center-village model Appreciative of the special situation in the Southeast, which fell outside the scope of the World Bank project, Bülent Ecevit announced the exclusion of the conflict areas from the village-town approach at the initiation of the Çavdar village-town at September 2, 2000. He announced that in ‘regions distant from the border in Southeast and East Anatolia’ they were ‘going down the route of Village-Townships,’ but that in ‘Southeast and East Anatolia, as a consequence of terrorist actions in recent years, many of our villagers have had to leave their homes. If they return now, they may encounter the curse of terrorism once again. Their economy will not easily be re-established. For that reason, in border areas, necessarily, we are going down the route of central villages’ (HRW 2002: 33). Ecevit gave GAP the assignment of making a study into the most appropriate framework for organizing settlement in the Southeast. Field research for the study was carried out between the February 19, and March 1, 2001. The master plan resulting from this study starts opened with the acknowledgement that the evacuation of villages and the displacement of people had inflicted great pains and sorrows on those involved, but that the forced evacuation had to be considered an opportunity for the creation of something new. Therefore, a plan for reconstructing the region should not be concerned merely with ‘return’, but with the creation of the conditions in which the ‘forced migrants’ could become more productive, both for themselves and for ‘their country’. The evacuation is regarded as an opportunity for the development of a new settlement structure that is more ‘rational’ and ‘vital’.17 Dispersed and unproductive settlement units were to be
17 The full (Turkish) text runs as follows: ‘Doğu ve Güneydoğu Anadolu Bölgesi Köye Dönü ve Rehabilitasyon Projesi Alt Bölge Geli me Planı” projesinin amacı, istemsiz
settling the southeast
109
changed into settlements units of sufficient size and potential (Oyan et al. 2001a: 1). In this narrative, the evacuation and destruction of rural settlements and the displacement of their inhabitants was transformed from an act of involuntary exclusion—both geographic exclusion and exclusion from social and economic networks—into a context that provided opportunities for the creation of something new. The aim of the plan was neither the restoration of property nor reparation of possessions of those whose houses and assets had been destroyed, but a more ‘exalted’, supra-individual objective—a rehabilitation of ‘the region’ (ibid.: 1, 7). It seemed as if past displacement and future development were only laterally related, linked in place and time but causally disconnected, which allowed for a separation of ‘past’ and ‘future’ as separate entities, with the present serving as a transitional phase. The jump from a war-induced displacement past towards a development-associated resettlement future put regional development first, and pushed the displaced to the margins.18 The plan disassociated itself both from a restoration of settlements and straightforward return of people, but also from the idea of a spatial concentration of the population in a limited number of settlements, either newly constructed or pre-existing. Although the researchers affirmed that it is not difficult to understand why a spatial concentration of the population would be deemed desirable by the governors and military, mainly because of their desire to supervise and control populations, they did not consider the spatial concentration of populations a viable option (ibid.: 11–12). First of all, it is emphasized that settlements in the area are not places for a random housing of individuals and family units, but the spatial expressions of kinship. For its inhabitants,
göçe maruz kalan ve bunun sorunlarını en iddetli ya ayan kesimleri, hem kendilerine hem de ülkeye çok daha yararlı olacak üretken bir konuma getirmek; istemsiz göçün topluma ve ekonomiye yüklediği maliyetleri, köye geri dönü ün doğru planlanması yoluyla bir fırsata dönü türmek ve bağlamda, geri dönü sürecini doğal akı ına bırakmak yerine, dağınık, ula ımı güç, hizmet götürme maliyeti yüksek, tarımsal faaliyetlere a ırı bağımlı yerle im dokusu yerine, yeni bir yerle im deseni olu turmak, tahrip olan konut ve kırsal hizmet altyapısını yeni bir anlayı la düzenlemek ve daha akılcı ve ya anabilir bir fiziksel ve sosyal çevre olu turmaktır.’ 18 The separation of time in two separate entities—the past and the future—related in time, but not in substance, is a remnant of the modernization approach in social theory, which identified the past with the domain of tradition (rural, agriculture, particular) and the future with the domain of the modern (urban, industry, universal). ‘Today’ was an in-between category, which, was supposed to achieve the conditions for modernity.
110
chapter three
place is not an arbitrary location in an abstract grid, which can be changed at will. Through generations, people come to feel attached to a particular place, which has become part of family, or tribal, history. Secondly, concentration of the population can run up against economic difficulties. The economy in the Southeast is agriculture-based and peasants are inclined to establish their house on or near their land. A concentration of the population implies the separation of farmers from their land, with negative affects on their livelihood. Thirdly, the resettlement in settlements other than those from which they were evacuated was met with opposition. A survey executed among evacuated 1,097 people from 297 selected villages revealed that around 90 percent of all respondents indicated that their desire to return home, to the settlements from they were evacuated, with even more, 98 percent, rejecting the proposal that they be resettled in settlement than other their own (Oyan et al. 2001a).19 The master plan for the Southeast was a development plan with rehabilitation as one of the key concepts (ibid.: 1). The concept of rehabilitation was used differently, however, than is the norm. In the international body of literature, the concept of rehabilitation is used in relation to actions that aim at a re-establishment of incomes, livelihoods, living and social systems (ADB 1998). Yet in the Village Return and Rehabilitation Development Plan, rehabilitation was defined as the rehabilitation of rural space and concretized, that is, realized as something physical—road and highway construction, construction of buildings. This new rural space would create the necessary framework for the establishment of State institutions (Oyan et al. 2001a: 5). The plan introduced two working-concepts: the sub-region (alt-bölge) and center-village (merkez-köy). The concept of a center-village was defined as a demographic-administrative project, comprising the concentrated settlement of people, but not necessarily into a single settlement unit. The center-village was preferred to the village-town, which also had an economic dimension, because the latter was considered too complex (among other things, entailing the establishment of rural industries, based on a division of labor between rural settlements) and therefore too costly. In Batman, Bingöl, Diyarbakir, Elazig, Hakkari, Mardin, Mu , Siirt, irnak, and Van the surveys gives identical results, with 93.6% of the respondents reported as wanting to return to their old settlements in each of these provinces. In Bitlis and Tunceli the number given is 85%. 19
settling the southeast
111
The concept of a sub-region was defined as a cluster of settlements distinguished from other settlements by economic, cultural, administrative or social characteristics. Supposed affinity and coherence between people and villages were used as characteristics to ‘border’ sub-regions. The center-village was defined as the settlement within a sub-region, which, by its characteristics—size, location, and infrastructure—could be turned into a junction or hub for the other settlements, and developed into an intermediate entity between the district town and small villages and hamlets. The center-village concept did not refer to the spatial concentration of people, but rather to the spatial concentration of services, and the articulation of surrounding settlements with this center.20 The concept of center-village was twinned to that of sub-region, the center-village as a nucleus, the sub-region as its surroundings. Organization The Ecevit government did not assign the coordination of the feasibility study and the implementation of return to village projects to either the General Directorate for Rural Services, KHGM, or the governors, but to the Regional Development Administration of the Southeast Anatolia Project BKI-GAP (Bölge Kalkınma Idaresi-Güneydoğu Anadolu Projesi). The choice of GAP was logical, made on the grounds of its administration, construction, and development capacity. First, GAP was a regional administrative body and therefore could coordinate the preparation and implementation of a program and the regional level. The GAP region included Adıyaman, Batman, Gaziantep, Kilis, Mardin, Siirt, ırnak, and Urfa, covering 75,358 km2, almost ten percent of Turkey’s total land area. Although the war-affected provinces of Bingöl, Bitlis, Elaziğ, Hakari, Mu , Tunceli, and Van were beyond GAP’s mandated territory, in practice this was not a major obstacle. GAP did conduct research for the feasibility study in provinces outside its mandated territory (and also conducted the feasibility study for the village-town in Van-Özalp).
20 The center-village concept was surrounded with controversy. Originally, it was used to identify settlements that could (potentially) perform center-functions (in providing public services, administration, etc) for surrounding settlements. During the course of the war, however, the concept was redefined, and came to mean a physical and social concentration of the population of small rural settlements into a nearby center.
112
chapter three
Second, GAP had considerable experience with large-scale infrastructure projects. GAP had originally been envisaged for the implementation of projects for hydraulic energy and irrigation, in which the construction was foreseen of 22 dams, 19 hydraulic power plants, and irrigation systems covering an area of 1.7 million hectares. Additionally, the GAP administration had experiences with resettlement projects (for the construction of eight dams in the region, 336 settlements had been or were to be evacuated, affecting, according to government statistics, about 181,000 people living in approximately 25,000 households, to rural or urban resettlements or else receiving compensation payments). Third, GAP was concerned with rural development issues (although GAP had initially been established as an infrastructure project, the mandate was changed in 1989). The concern with ‘rural development’ was in fact a response to the emergence of the PKK. Development of the region was supposed to contribute to the struggle against what the authorities referred to as separatism and terrorism. The government hoped that rural development would help to minimize the number of PKK supporters (White 2000: 111). In addition to the objective of rural development, the subject of ‘social development’ entered the agenda as a consequence of the government’s search for a more thorough understanding of the process by which millions of ‘mountain-Turks’ had turned into Kurds (Akin 1993; Özok 2004: 4). While rural development was supposed to contribute to the development of a market oriented, capital-intensive agriculture, social development was supposed to contribute to abolition of the extended family and tribal institutions, and to the dissemination of a modern lifestyle, characterized by emergence of nuclear families and reliance on the state instead of tribe for security and resources. The modern lifestyle identified with development and loyalty to the state was thought to be the best possible weapon against the existence and persistence of a Kurdish identity, regarded as form of pre-modern tribalism. The KHGM and the governors had also had experience with large projects, but had obvious disadvantages. The KHGM could execute projects at the regional level and it was a strong in construction, but it was weak in development. The governors could coordinate projects at the provincial level, but not at the regional level since the abolition of the State of Emergency (OHAL) and the State of Emergency Regional Governor. Additionally, the governors had proved to be unwilling to execute projects in areas that had been evacuated. Taking all this into consideration, GAP emerged as the most appropriate institution for the coordination of the study and the initiation of projects. However,
settling the southeast
113
this did not imply that the governors were excluded. As the highest provincial authority, the governors were to be involved in the coordination of projects in their respective provinces. The Pilots The Village Return and Rehabilitation Development Plan proposed twelve pilot projects, one pilot per province (see Table 3.2). Table 3.2 Center-village pilot projects (Sources: Oyan et al. 2001b, c, d, e, f, g, h, i, j, k, l, m; Meclis Ara tırma Komisyonu 1997; Doğanay 1993) Province
District
Batman
Gercü
Vergili
Bingöl
Genç
Yağızca
Bitlis
Center
Karınca
Diyarbakir
Çüngü
Yeniköy
Elaziğ
Palu
Arındık
Hakkari
Center
Kaval
Mardin
Ömerli
Kocasırt
Mu
Center
Üçevler
Siirt
Center
Sağırsu
8 villages
Cizre
Sırtköy
Tunceli
Mazgırt
Bulgurcular
Van
Gürpinar
Yalınca
4 villages, 9 hamlets 4 villages, 2 hamlets 6 villages, 36 hamlets
ırnak
21
CenterVillage21
Sub-Region Number of dependent settlements 13 villages, 4 hamlets 3 villages, 22 hamlets 6 villages 9 villages, 3 hamlets 2 villages, 6 hamlets 6 villages, 4 hamlets 7 villages, 1 hamlets 9 villages, 18 hamlets
See Annex 1 for maps of the pilot projects.
Date of Nomination as Center-Village and Nominating Institute 1983 Council of Ministers 1997 Council of Ministers 1983 Council of Ministers 1983 Council of Ministers –
Budget (in billion TL) for implementation of the pilot 5,403 5,279.5 2,493 2,973.5 4,746
–
11,750.5
–
5,122.6
1995 State of Emergency Regional Governor 1983 Council of Ministers 1983 Council of Ministers 1983 Council of Ministers 1995 State of Emergency Regional Governor
5,972 3,129.5 3,619.5 3,694 6,703.5
114
chapter three
The Village Return and Rehabilitation Development Plan proposed 12 pilot projects, one pilot per province. The pilot projects were not about 12 settlements, but about 12 sub-regions, each of them including one center-village and a number of dependent villages and hamlets, varying in number between 6 and 42. The total number of settlements involved was 12 center-villages, 77 villages and 105 hamlets, or 194 all told. The pilot plan did not mean that 194 depopulated villages would be reconstructed, for not all the settlements included in the pilots were depopulated settlements. In one case—the settlements that were part of the Yeniköy pilot project in the Çüngüs District of Diyarbakir province—the villages and hamlets were all inhabited. As the government study explains (Oyan, Ersoy et al. 2001: 7): Just since the end of the 1980s and in particular since the 1990s, for reasons of terror, village evacuation in East and Southeast Anatolia occurred on a wide scale. It is necessary to give priority to the issue [in these areas]. The solution to the problem is not to stay within the boundaries of a return to the evacuated villages. . . . Villages that have not been evacuated at all have to be included in the alternative models that are part of the framework of the sub-regional development plan.
Other pilot projects, such as those in Batman, Bingöl, Bitlis, Mu and Van, were composed of both depopulated and inhabited settlements. Villages inhabited by village guards recruited during the original scorched-earth campaign were also included in the pilot projects. It would appear that prospective returnees were to be settled next to village guards, as far as information is available—leastways, this is the case with the projects in the provinces of Batman, Bingöl, Bitlis, Hakkari, Mu , ırnak and Van.22 Finally, not all the ‘evacuated’ settlements were eligible for repopulation. The pilot in the Yağızca sub-region in the Genç district of Bingöl showed that some of the evacuated settlements (Küçükbayırlı and Bayırlı) were not to be reconstructed: their populations were to be resettled at locations more suitable to the State (see Map 1).
22 In three out of the twelve provincial research reports the additional income from koruculuk is quantified: in the sub-region Kaval, in Hakkari, 9.1 percent of the population has additional income from koruculuk, in the sub-region Yalınca, in Van, it is 9.3 percent, and in the sub-region Sırtköy, in ırnak, 12.8 percent.
settling the southeast
115
Box 3.1 Destruction, Reconstruction and Legislation In the course of military operation, rural settlements were evacuated and destroyed on a large scale. Occasionally, major towns were also targeted. In 1992 the provincial town of ırnak was nearly obliterated by Turkish security forces. Turkish authorities stated that the PKK had attacked the town, provoking a military reaction, although ırnak residents denied that claim. Homes and businesses were so badly damaged and the residents so fearful that only between 2,000 and 3,000 people remained in a town that had contained 35,000 inhabitants (Helsinki Watch 1992). That same year, the Turkish military also attacked and nearly destroyed Çukurca, a district-town in the province of Hakkari. The near destruction of the two towns was not in itself a military objective. On the contrary, the towns were appropriate urban pockets for spatial concentration, of both inhabitants and state security forces, and the establishment of control in a predominantly rural and mountainous region, where guerrilla forces were able, at the time, to move around with relative ease. Therefore the authorities put significant efforts into the reconstruction of both ırnak and Çukurca. For reconstruction purposes, the Turkish parliament amended the Disaster Act No. 7269 with a law numbered 3838 in 1992. This law gave the authorities the means to release resources for the repair or reconstruction of damaged or demolished houses and other properties. The aim of reconstruction was to prevent a permanent, collective migration from the two devastated towns and encourage return. An exodus of civilians from the two towns would have been a problem for a successful counter-insurgency, as it would imply the loss of urban pockets where state security forces could establish control. Such an amendment to the Disaster Act had not been considered when a return to rural settlements first entered the political agenda. Since 1997, however, different governments have considered new drafts of Settlement Act No. 2510, a law that was originally passed in 1934 and had achieved notoriety when it was used to organize the forced resettlement of Kurds in the 1980s and 1990s, to settle Kurdish nomads23 and to resettle populations affected by the construction of dams.24 It was against this backdrop that feasibility studies were carried out in 2003 for the establishment of a center-village project in Çüngüs and Return and Rehabilitation Projects in Buçuktepe (Diyarbakir Center), İnkaya (Kulp District of Diyarbakir) and Dibek (Lice District of Diyarbakir).
23 The settlements of the Beritan people in the villages of Alibeyköy, Tepekonak, and Çeltikli (in the Bismil district of Diyarbakir), the Kosan in Tepekonak and Çeltikli, and the Alikan in Çölgüzeli (Siirt) were all carried out under the provisions of the Settlement Act 2510. 24 This includes villages affected by the construction of the Atatürk, Dicle, and Batman Dams.
116
chapter three
A revision of the law was deemed necessary because many revisions (17 amendments since 1934) had turned it into a regulatory patchwork lacking any real cohesion. An integrated, new version of the law was considered desirable in anticipation of the implementation of large-scale settlement projects in the war-affected Southeast, in particular since the authorities were planning to develop a new settlement structure (rather than merely revamp the old one of village-towns). The draft of the revision clearly anticipates the intended settlement structure changes (and later proposed in the Village Return and Rehabilitation Development Plan). Article 16 of the new draft states that ‘for the purposes of a physical organization of settlements in rural areas (. . .) villages without prospects’ (i.e. settlements whose location was deemed inconvenient or inappropriate, for reasons of economics or security) might be ‘moved to more convenient settlement locations.’ The new law also shifted implementation responsibilities from an interdepartmental settlement commission to the National Security Council, MGK (Milli Güvenlik Konseyi), formally an advisory board established by the military after their coup of 1960, but until a reform in 2004 de-facto the highest state-institution (see Chapter 4). The institutional responsibilities are explicated in Article 13, which states that the Council of Ministers was to decide on evacuations ‘for reasons of national security’ in accordance with recommendations made by the National Security Council, after which the Office of the Prime Minister was to indicate a settlement alternative. The draft law continues with Article 14 thus: ‘. . . if [the displaced] decline to live in the places directed by the Office of the Prime Minister, their rights [to resettlement] will be annulled by the local settlement commission. Families in this situation may not make a second housing application’—which effectively turns the law into legislation for forced resettlement. The draft law was sent to Parliament by three subsequent Prime Ministers (Mesut Yılmaz of the Motherland Party, ANAP, on May 10, 1997, Bülent Ecevit of the Democratic Left Party, DSP, on July 5, 1999, and Abdullah Gül of the Justice and Development Party, AKP, on January 7, 2003), putting return-to-village projects firmly on the agenda. As of the end of 2006, however, the proposed legislation still had not been put to a vote.
The pilot projects were not implemented. Mainly, this was because of institutional backlash, the overt and covert opposition and rejection of state-institutions to projects decided upon by the Office of the Prime Minister. No information is available about the reaction of the army to the master plan for a return to villages, but, as discussed above, the preference of the military generally was for settlement in urban or urban-like settlements, where population is concentrated. Information is available, however, regarding the response of the governors. From the start, the governors in the region (including the regional State of Emergency Governor, who remained in office until the State of Emer-
settling the southeast
117
gency was lifted in 2002) had preferred a study that would result into the classification of the evacuated rural settlements into two types: those settlements which, mainly for reasons of security, were not deemed appropriate for reconstruction, and those settlements which were appropriate for reconstruction because of the feasibility of turning them into centers for the spatial concentration of populations, appropriate for supervision and control of the population. The researchers however, kept to an approach of articulated settlement, distancing themselves from the governors’ proposal to concentrate the population into a limited number of settlements for the reasons discussed above (people’s emotional attachment to their villages, the likely economic difficulties of spatial concentration due to its separation of villagers from their land, and the rejection of concentration by the villagers). There thus developed, it would appear, something of an impasse between the two positions, an institutional conflict and power battle between the appointed governors opposed to the plan (with implicit military backing) and the Regional Development Administration of GAP. Ultimately, some of the governors of provinces openly prevented implementation of parts of the program, on the basis of security concerns. The governors in question were those of Bingöl, Hakkari, Siirt, and Van (Oyan et al. 2001d, g, j, m). The Governor of Bingöl rejected the inclusion of the settlements Servi-Sarıbudak in the pilot. The Governor of Hakkari only allowed return if resettlement was accompanied by the opening of police stations. The Governor of Van did not consider re-population appropriate to six of the settlements included in the Yalınca pilot in his province.25 The Governor of Siirt stated that the dispersed settlement pattern was the major problem in the war against terror and subsequently refused to give permission for a pilot project intended to effect a return to nine evacuated villages in the Bağgöze valley in the Eruh district. Further to this, the governor then stipulated in a letter to the GAP Regional Development Administration that settlements should be within a radius of five kilometers of a police station (Oyan et al. 2001j). It should be noted that the governors were not all of one mind on this. The Governor of ırnak, interestingly, proposed a pilot in Sırtköy, assessing its development potential to be better than that of Düzova, and in spite of the fact that, it was said, security could be more easily
25
These villages were: Sımali, Dokuzdam, Cepkenli, Topyıldız, Dikbıyık and Keçilioba.
118
chapter three
in provided Düzova, with the Cizre-İdil military base close by (Oyan et al. 2001j). Nevertheless, the governors did not implement the pilot projects for the settlements proposed in the master plan, and persisted instead in their desire to establish concentrated settlements, such as the İslamköy project in the Kulp district of Diyarbakir. Conclusions The basic argument of this chapter is that the two so-called ‘returnto-village’ models, those of the village-town and center-village, were essentially concerned with a re-construction of rural space, each developing an articulated and integrated settlement structure. When used, the concept of ‘rehabilitation’ did not refer to a rehabilitation of the displaced (by means of a recovery of their livelihood), but to the treatment of structural disabilities in the settlement structure of the region: the many small rural settlements, their dispersed distribution, and lack of inter-settlement articulation. The center-village model, laid down in the Village Return and Rehabilitation Development Plan, straightforwardly considers the evacuations of small rural settlements as an opportunity for designing a more promising (integrated, productive) settlement structure. Both of the models, village-town and center-village, were considered as tools for the development of a new settlement structure, eventuating in a modernization of the region. Neither of the models was implemented, primarily because of institutional opposition, both national and international. Implementation of the village-town model was seriously hampered when the World Bank decided not to release loans for its implementation in areas where internal-displacement as a result of war had occurred. Furthermore, and more importantly perhaps, plans for a reconstruction of rural space was regarded with great suspicion by the military and the local administrators (governors and district-officers), who were reluctant to move forward with such a plan for fear that re-population of the evacuated areas might contribute not to the development of the Southeast, but rather to the restoration of a supportive environment of the PKK.
CHAPTER FOUR
THE INTEGRATED SETTLEMENT NETWORK Towards a New Spatial Organization of the Countryside Up to now we know of no city which has rural characteristics. But the converse, a village possessing elements of the town, is feasible. Haim Halperin A central issue I wish to address is the rural-urban divide that pits cities against the countryside as two irreconcilable social, moral and physical formations. Since the early decades of the nineteenth century, overcoming the rural urban divide has been a persisting but largely unrealized dream. John Friedmann1
Introduction Having considered the resettlement practices in the areas affected by the war between the Turkish Armed Forces (TAF) and the Kurdistan Workers Party (PKK) we now turn to the genealogy of some key concepts: abolition and concentration of villages, village-towns and center-villages. This discussion links the ‘(re)settlement issue’ in the Southeast to the ‘settlement issue’ in Turkey in general, and adds to it an historical perspective. The genealogy of concepts will reveal a quest of bureaucrats, politicians and intellectuals for a new type of settlement entity and structure in the countryside. The organization of space by means of the establishment of a new type of settlement and an articulated structure was supposed to consign the irregular dispersal of small settlement entities to history. Regardless of the details of the new entities and their concomitant structures, they were all supposed to stimulate a trivalent process consisting of the down-scaling of administration (state-building), the fostering economic integration (through a refined division of labor), and the creation of a common cultural basis for social life (nation-building). This chapter will discuss some of the intellectual foundations of this quest for a new settlement type and 1
Haim Halperin 1963: 7 and John Friedmann 1996: 129.
120
chapter four
structure, considering the issues of rural settlement and administration, and looking at the proposals for a reorganization of space (under such names as center-village and village-town). Modern Spaces Philosophically, modern space is conceptualized as uniform and continuous in all directions. Sociologically, modern space is characterized by the domination of industrial and urban forms. Urban-based industrialization has been considered to be the driving force of the modernization of society. Although many accepted the coincidence of urbanization and industrialization as a law of history, the idea has not gone unchallenged. Different thinkers (Howard 1902; Kropotkin 1912; Köymen 1937; Mumford 1961; Halperin 1965; Friedmann 1996) have criticized this spatial representation of modernization, arguing that the coincidence of these two processes (urbanization and industrialization) was not an unfolding law of history, but its gravest mistake. In Turkey, this critique of the classical conceptualization of modernization as a simultaneous process of industrialization and urbanization eventuated in the fermentation of new ideas on the making of an original settlements entity with a new economic foundation, a settlement entity that was beyond the modern divides of rural and urban, and agricultural and industrial. In the 1950s and 1960s such ideas were inspired by the work of the spatial planner, John Friedmann, and the agrarian economist, Haim Halperin. Both Friedman and Halperin argued that the urbanization of industry creates cities and villages as irreconcilable social formations, fostering the former, draining the latter, and having a tremendous negative impact on society (Friedmann 1996: 129–30; Halperin 1963: 1–9). Halperin in particular was influential on the Turkish social scientists (rural sociologists) who had put efforts into re-thinking the rural-urban and agricultural-industrial divides (Koçtürk 1967; Oğuz 1976; Tütengil 1975). The book in which Halperin proposed the development of a new type of settlement, one founded upon the integration of agriculture and industry, was translated into Turkish in 1965 (Halperin 1963; Halperin 1965) and elaborated upon by the nutritionist Osman Nuri Koçtürk. Koçtürk, who graduated at the Military Veterinary Academy, was adviser to the (İnönü) government in 1963 and initially much concerned with what he called food-colonialism, which he defined as the assimilation of Turks by the new empires of
the integrated settlement network
121
modern (American) food habits and diets (Koçtürk 1966). Koçtürk thought that the independence of the Turkish nation needed to be safeguarded against this ever-growing menace with the development of agro-industrial centers in rural areas, fortresses which would foster social, cultural and economic activities to support national life (Koçtürk 1966). Introduced by Koçtürk (1967), the concept of tarsan, which could be translated into English as agrindus, was derived from the components tarim (agriculture) and sanayi (industry). It bears a close resemblance to the concept of the village-town, as explained below. Halperin (1963: 7) deemed it necessary to create new forms of human settlements, which he envisaged as industrial centers in a rural setting. Historically, he argued, this search for intermediate forms was not new, but its realization had remained incomplete. Among the ideas that Halperin refers to was the proposal during the early years of the existence of the Soviet Union to create agrogorods. These agricultural towns were envisaged as vast collectives (kolhozen). Although the notion was not taken any further and the idea itself remained somewhat unclear, Halperin recognized in it a quest for the making of a new spatial formations be based on an economic integration of agriculture and industry. Thus it was that he preferred to name the new spatial formation tarsan (agrindus), etymologically derived from its two dominant roots. Halperin’s agrindus envisaged not only an integration of agriculture and industry, but also the formation of a new settlement type, for which the industrialization of agriculture formed the material base. Basically, the agrindus synthesizes the rural and the urban and the agricultural and the industrial: Ours is a bewildered generation. We have scaled the heights but it seems that the ground is slipping from beneath our feet (. . .) (T)he village is beginning to vanish from the face of the earth (. . .) Are we so certain that no crisis is in store for industry and that we have not gone beyond the bound of prudence in concentrating millions of human beings in single cities (. . .). It is difficult, if not completely impossible, for any man in our age to imagine the future city with teeming millions. Can it possibly be an integrated unit? (. . .) But why should we give up rural society entirely? Are we indeed capable of developing a better, more integrative society; one that is more robust—not only physically but also morally—more patriotic and more upright? (. . .) Agriculture can be combined with industry without undermining the age-old social asset—the village. We can improve and even reform the village and bring it into line with changing conditions (. . .) Up to this day we know of no city which has rural characteristics. But the converse, a village possessing elements of the town, is feasible. Moreover, it is far more desirable to establish an industrial center in a rural setting. Within a group of 25, 30 or even 40 villages, let us say, an
122
chapter four industrial setting would be developed possessing all those attributes of a city, which the villagers find so attractive. The urge towards mobility, so deeply rooted within us, would also be satisfied. (Halperin 1963: 4–5).
Friedmann’s model of spatial integration seeks to transcend the ruralurban and agriculture-industry divides. He uses the terms agropolitan development (Friedmann 1988) and modular urbanization (Friedmann 1996) to describe an alternative development model going beyond these divides. The basic idea is that urbanization is not only to be understood as the migration from village to city, but that it is possible also to consider urbanization and industrialization in terms of components or modules, and that new settlement entities can, and should, be created by the implantation of urban and industrial modules into the countryside (Friedmann 1973: 24, 54). Friedmann describes social change associated with spatial integration as resolving into four processes: i) an increase in the level of transactions between territorial divisions which (supposedly) lead to a more complex territorial division of labor, ii) an increase of State control, iii) extension of a common basis for social life, a shared culture, and iv) an increase in migration from rural areas to various centers (ibid.: 1973: 24, 68, 70). It is assumed that the ‘traditional’, selfcontained space is gradually dissolving and being replaced by a modern integrated space. Friedmann’s model of spatial development was used by the Ministry of Village Affairs and Cooperatives in the explanation of the modernizing effects of the village-town project implemented in Özalp-Dorutay in 1978. The Rural Settlement Issue In 1982, the State Planning Organization, DPT (Devlet Planlama Te kilatı), published a study ranking settlements in Turkey, using a classification system of settlements based on the functions performed (administrative, economic, social, cultural, political) and their spatial impact (local, sub-regional, regional, national, and international). The study was to contribute to a constructive policy in which settlements would be linked in a vertical network, from village group centers connecting the rural grid to regional and national centers and ultimately the international centers joining Turkey to the outside world. This tying together of settlements may be defined as the way in which the state, as a spatial construct (the capital), establishes itself as the center of society.
the integrated settlement network
123
The idea of categorizing and ranking settlements was not new in Turkey, but its all-embracing implementation was. The study was executed state-wide and revealed that Turkey had only one international center, İstanbul, four national centers (Adana, Ankara, Gaziantep, and İzmir), and eleven regional centers (Bursa, Diyarbakir, Elaziğ, Erzurum, Eski ehir, Kayseri, Konya, Malatya, Samsun, Sivas, and Trabzon). The other centers were classified as sub-regional centers, small town centers, and village group centers (the idea of center-villages wais based on the category of village group center). Table 4.1 The ranking of settlements in Turkey (Source: DPT 1982) Rank
International Center National Center Regional Center Sub-Regional center Small Town center Village Group center Village
Number
Average size of the area influenced (km²)
Average diameter of the area influenced (km)
Average Statistical number of average settlements influenced
7
1
780,576
500
4
4
6
4
156,115
223
2.20
2.75
5
11
15,530
125
3.62
5.27
4
58
10,548
58
6.82
8.71
3
505
1,348
21
2.20
2.65
2
1,337
407
11
18.33
26.28
1
35,132
21
2
–
The DPT study revealed a particular weakness of Turkey’s spatial system. Turkey had relatively few centers which connected the rural grid to the sub-regional, regional and national centers, particularly when taking into account the large number of uncategorized and unranked hamlets, a number probably greater than that of villages, the lowest ranked category of settlements). In fact, the number of settlements in category 1 and thus dependent on or ‘influenced by’ those of category 2 is expressed in relative terms at over 70:1 (compared to five or so sub-regional centers dependent on each regional center, or 5:1). The high average number of settlements dependent on these village group centers indicates a thin distribution of centers (or high level of settlement scatter) in the countryside. This idea of weakness may be illustrated by
124
chapter four
comparing the last two columns in Table 4.1, the average numbers of settlements influenced by a center, and the statistical averages. Only at the level of the international center are the actual average number of settlements influenced and the statistical average the same. At the other levels, the actual influence is lower than the statistical average, indicating a weak integration of settlements in the space of the nation-state, in particular (but not exclusively) at the level of villages. The methodology and results of this ranking appear somewhat questionable, particularly in respect of the validity of the settlements influenced and area of influence columns—certainly there must be a high degree of subjectivity in the various weightings presumably afforded to the various criteria factored in. One may ask, for example, what it means to arrive at a calculation for Istanbul, the international center, that affords it an area of influence spanning (exactly) 500km, or how it was that the international-to-national settlements influenced average arrived at was precisely the same as the statistical average. The pyramidal structure is an idealized modeling rather than an accurate reflection of reality (whereas administrative hierarchies may be strict, this is not the case for other, socio-economic and cultural, functions), suggesting a measure of artifice, particularly at the upper end of the scale. The important point here, however, is that the state employed a centralized model which recognized a great weakness in terms of integration at the local, rural-to-urban level. The number of villages influenced by the small town centers compared to the statistical ratio (of villages to small town centers) showed a higher discrepancy (a ‘shortfall’ of almost a third) than any of the other ranking category proportions. This perception by the state of its own workings, or failings, had profound implications for policy, suggesting the necessity for major reform in order to further integrate the system, draw the rural grid more closely into the state system and continue the nationalist state project initiated with the republic. Indeed, it was this perception that lay behind the center-village and village-town ideas that became key concepts in the 1970s and 80s, and were later revived in the 1990s when return-to-village came onto the agenda. Notwithstanding the extent of any problems there might have been in the way that figures were calculated in the DPT study, or what it was exactly that was being calculated even, these were not its greatest weakness. This was the glaring omission of hamlets, the category zero in more than one sense. Indeed, the total number of all the settlements counted in the study was probably smaller than the number of
the integrated settlement network
125
settlements omitted, for this reason. Striking, not to say incredible, is the absence of reference in the system of classification and ranking to the tens of thousands of hamlets dotted throughout the nation, as if there are no words to name them. It calls to mind the idea of ‘negative villages’ (‘menfi’ or ‘negatif köyleri’), a term used by the architect Abdullah Ziya to depict settlements ‘without value’ and being ‘not worth describing’ (‘negatif kölerin ekli ve bedii bir kiymeti olmadığından fazla bahse lüzum görümyoru’) (Ziya 1933: 374). Categorized as negative villages, these settlements were denied existence in the legitimate discourse (and therefore not described), their inhabitants likened to dead bodies (‘I felt as if I had stepped on a grave’) or to non-humans (‘bodies wiggling in the cavities of the slope’ (Nalbantoğlu 1997: 205–206). The Rural Settlement Grid The study reflects the concept of modern space as a unified whole of interrelated and hierarchically organized elements, to be created by a process of spatial integration, and defined as the multiplication of dependency relations between settlements.2 The modern settlement system represented in the DPT study is composed of four settlement types: metropolis, city, town and village (hamlets, of course, going unmentioned). The settlement type metropolis (büyük ehir) specifies a settlement that has a population of more than 250,000; cities ( ehir) are defined as settlements with a population of under 250,000 but over 20,000; while towns (kasaba) are settlements with between 2,000 and 20,000 inhabitants—which is where the rural-urban divide comes in, cutting through the category of towns, because towns with a population above 10,000 are termed urban, while towns with a population below 10,000 are termed rural.3 As for the settlement named village (köy), this
Indeed, Friedmann’s work is inspired by that of Neil Smelser, who looked upon modernization as a process of ‘structural differentiation’ and integration. For Smelser, this process of differentiation and integration denoted the move from a ‘traditional’ society, where integration is bound up with kinship status or tribal membership, to a ‘modern society’ which is integrated through political and cultural networks (Long 1977). 3 It is hold that ‘urbanity’ and ‘rurality’ are not only a matter size graduation, but also of structural differentiation. A primary (agrarian) economy makes a settlement rural, while a secondary (industrial) or tertiary (service) economy makes a settlement urban. This means that a town of 5,000 inhabitants, with its population employed in tourism, is said to be more urban than a town of 10,000 inhabitants, with its population employed in animal husbandry (Özgür 2002). 2
126
chapter four
is defined by The Village Act (Köy Kanunu), passed by Parliament on March 18, 1924, which recognizes a village not only in terms of its size (150–2,000 inhabitants), but also in structural terms, as a settlement with its ‘people living compactly or dispersed, who jointly own a mosque, a school, a pasture, a summer pasture and part of a forest reserved for fire-wood form a village, together with their vineyards, gardens, and fields’ (Village Act, Articles 1, 2, and 89) (Resmi Gazete 1924). That Turkey’s tens of thousands of hamlets went unrecognized in the DPT study, like Ziya’s ‘negative villages’, is quite remarkable really. Essentially, the hamlet was not considered an appropriate settlement type for a modern society, but a (mere) remnant from a traditional (outmoded) past. With its view to a modern settlement system into which Turkey should aim to be transformed, the study implicitly passed judgment over the value of hamlets—they had none. Nevertheless, for the purposes of state administration, hamlets had to be categorized, but this turned out to be somewhat problematic. The Village Acts defined hamlets in terms of size, as settlements with a population of less than 150. The State Institute of Statistics, however, which keeps the records for the numbers of villages and hamlets in Turkey, uses an upper limit of 250 inhabitants. Meanwhile, in a decree issued on November 16, 1987, the Minister of Internal affairs determined the minimum population of a village to be 500 people, although the 150 minimum figure could be employed, but only where the Minister considered it appropriate. The same decree also stated that to qualify for the status of a village, a settlement had to be at least five kilometers distant from the nearest town or village, thereby adding a social dimension of spatial positioning to that of population size.4 Yet ten years later, the same ministry proposed in a draft law for the reform of local administration to determine the minimum population of a village at 300 people, and to end village status for settlements with fewer inhabitants—but in the final draft of the law, the provision was struck out (Özefe 2001a: 82). Hamlets, as rural settlements, also have their lower limit, defined according to interpretations of laws and directives in terms of population size, physical size (span) and number of houses. To be counted as a rural settlement, the
4 The Ministry could deviate from this rule and give special permission for the establishment of a village within five kilometers of the nearest town/village in cases of particular social or cultural conditions, safety or public security concerns, barriers (infrastructure or natural) like dams, rivers, etc, preventing access, or for reasons of public services.
the integrated settlement network
127
inhabited space must have a radius of at least 150 meters; also, there must be a population of least 50 people, or, alternatively, the settlement must be composed of at least five houses with a minimum of 30 inhabitants. If an inhabited area does not conform to these standards it is mot considered to be a ‘settlement’ but a ‘settlement core’ (‘iskan çekirdeği’) (Güzelsu 1983b: 7, 10). The rural settlements classified as hamlets are defined not only by size (however measured, by population, houses or span), but also by regional, socio-economic and characteristics. We saw this indicated above in the spatial positioning specification referred. Definition by state organizations represents a ‘top-down’ approach to settlement; a ‘bottom-up’ approach assumes the cultural facts as starting points, most obviously expressed in language. The language of settlement is very rich in Anatolia, especially in respect of hamlets, which are referred to by such names as mezra, çiftlik, oba, yayla, kom, divan, and mahalle. These names reflect a system of classification not based on size, but on socioeconomic characteristics and regionality. The mezra for example is a permanent settlement of a somewhat dispersed type. Literally, mezra means arable field. Its economic foundation consists of arable farming, even though animal husbandry might be an important source of income also. The mezra is a common settlement type in Turkey, including the Southeast. Because it is such a common settlement type, the concept mezra is also used in government statistics to refer to the category of small rural settlements in general, carrying the meaning of the English word hamlet. In Ottoman surveys, the concept of mezra was used in a similar way. It was simply considered a cultivated but only periodically inhabited area. A çiftlik, literally meaning farm, is another settlement type based on arable farming. The çiftlik provides a livelihood for several families and consists of a couple of houses established on a large land estate, which makes the çiftlik a smaller version of the mezra. This settlement type is widespread in the plains and lower parts of the country (Güzelsu 1983b: 5–8; Tunçdilek 1974: 55–61). The oba is a settlement type widely diffused in mountainous areas and based on animal husbandry. Oba literally means ‘large nomad tent’ or ‘nomad group’. The settlement type is established on summer pastures and associated with semi-nomadic animal husbandry production. The oba used to be a semi-permanent settlement type, inhabited in summer, and left behind in winter. However, some oba have been turned in permanent settlement units, under influence of more intensive land use for example. A kom is also a settlement type based on animal husbandry,
128
chapter four
but it is said to be smaller than the oba. However, oba is of Turkish origin and kom might be of Kurdish origin. The yayla is another name for a seasonally inhabited settlement associated with animal husbandry (summer grazing), and common in mountainous areas of the Southeast, (like oba, but without the nomadic connotations, and instead with connotations of a summer retreat).5 Strictly speaking the divan is not a settlement type, but an administrative body, established in the Tanzimat period of the Ottoman Empire. As an administrative body, the divan was established after the administrative clustering of rural settlements in a nahiye, to be defined in today’s terms as a sub-district center. In the northern part of western Anatolia, the İstanbul-Bolu-Samsun triangle, the term divan is also used to designate small rural settlements of a compact kind. The word mahalle is used as an administrative term also, and to designate a city neighborhood; it has another use, however, that of referring to a settlement in transition from settlement core and hamlet(s) to village. Somewhat teleologically, the mahalle is regarded as a ‘village in the making’, a village which comes into existence by means of a multiplication and unification of settlement cores or groups of houses. Interestingly perhaps, the idea that a reverse process could occur, from village to hamlet, is not recognized in the Anatolian taxonomy of settlement names.6 Number and development of the small rural settlements The large number of villages and hamlets was a matter of great concern to bureaucrats, politicians, and developmentalists (their decline or persistence was considered an indication of the success or failure of the modernization process). Given this concern, one might expect statistics to have been widely evaluated and discussed. This seems not to have been the case, however. On the contrary, statistics are difficult to obtain, and if available, have to be considered with care due to a lack in clarity in the definition of the units counted. Table 4.2 presents statistics from various (secondary) sources. Exercising due care in 5 Yayla has also a second meaning referring to a high mountain area/plateau near the sea with high humidity/rainfall, but this being a geographical term rather than settlement type. 6 This ‘reverse development’, from village to hamlet, has been seen, for example, in the case of Mira, discussed in Chapter 6. Another form of ‘reverse development’, the phenomenon of once permanently inhabited settlements becoming seasonally inhabited, significantly so at least, is also considered in Chapter 6.
the integrated settlement network
129
comparing and interpreting these numbers we can nevertheless draw three main conclusions. First, the number of rural settlement is high, in particular compared to the total number of settlements. Second, the ‘smaller’ (population less than 150–250) rural settlements (i.e. hamlets) tend to outnumber the ‘larger’ ones (population more than 150–250). Third, the number of rural settlements grew during the this perid, but the rise in the number of hamlets was greater than that of villages (both proportionately and in absolute terms—in fact the hamlet number seems to have overtaken the village number during the second half of the twentieth century). Fourth, the proportion of people living on the countryside reduced considerable (from 73% of the national population in 1950 to 35% in 2000), but their absolute number increased. Table 4.2 Population, villages, and hamlets in Turkey (Sources: Tütengil 1975; Güzelsu 1983; Dülger 1984; Doğanay 1993; DPT Ihtisas Raporu 2000, DIE Genel Nüfus Sayimi 2000) Population Rural Rural Number Number Total Total (million) population population of of number of number of (million) (%) Villages Hamlets rural settlements settlements 1950 1970 1985 2000
19 36 51 67
14 22
73 62
23
35
34,063 35,400 36,155 35,427
30,198 38,600 52,398 37,573
64,261 74,000 88,553 72,990
64,825 74,934 89,076 75,860
These statistics on the number of settlements seems to contradict the dominant idea in Turkish social sciences that the small rural settlements are bygones from an earlier era, with their ultimate disappearance being just a matter of time and the task of policy being to speed up the demise. Also, historical studies reveal that the small rural settlements are not necessarily remnants from the past at all, as many were formed in various modern situations. Hütteroth shows that until the middle of the 19th century there had been no new settlement of note in the greater part of Anatolia for about 350 years, and derelict villages and abandoned fields dominated. During the last decades of the 19th century, however, conditions for the colonization of land for permanent settlement improved as a result of increased control of the army over nomadic tribes, the security of property, and the development of transport (the railway system) connecting land to markets. The settlement of mühacir—migrants from lost territories of the Ottoman Empire—also
130
chapter four
impacted on the settlement pattern. Indigenous peasants lost parts of their traditional pasture land to mühacir settlements and tried to secure the rest by more intensive usage. After the arrival of the migrants, peasants started to use their summer pastures in the mountains more intensively. Some of these pastures even developed into settlements in their own right (Hütteroth 1974: 21–7). Tunçdilek (1974: 51) enumerates five (interrelated) factors, which also give an idea of the various factors underlying modern settlement (hamlet) formation. First, social systems depending on patriarchal family organization lost importance. The desire to live independently led to an increasing number of newly-formed small settlements. Second, disputes and feuds between relatives led family members to abandon their homes and establish new settlements. Third, population growth resulted in an increase in the number of households, the basis of rural settlement patterns and economy, and thus to a rise in the number of settlements. Fourth, numerical increase in family members and diminishing subsistence led to the colonization of new land, and consequent settlement establishment. And fifth, families squatted on governmentowned land and forest areas in order to benefit from the legal possibility of acquiring ownership rights. A dispersion of settlements and the formation of hamlets, or settlement cores, importantly took place in reaction and relation to modern developments (population growth, industrialization, private land-acquisition opportunities). These settlement entities are clearly not to be located in history in the sense of things past, as though of little contemporary significance. Dispersed rural settlement structure The rural settlement structure is characterized by a large number of small settlements (Table 4.3) and a high level of dispersion. This dispersion has three dimensions. First, rural settlements are irregularly distributed in space. They form a thick scatter in some areas and thin in others. For example, in the Marmara (İstanbul) region there is an average of one settlement for every nine square kilometers, while in Hakkari each settlement occupies an average of 77 square kilometers (figures that obviously refer primarily to rural settlements, given their overwhelming preponderance). (Tunçdilek 1974: 50). Furthermore, the population density in particular areas is rather thin (such as in Van and Hakkari, and provinces in East-Anatolia Region: see Graph 4.1).
Graph 4.1 Population density (Source: State Institute of Statistics)
the integrated settlement network 131
132
chapter four
Second, this irregular spatial distribution of rural settlements itself conceals a further irregularity, within the category of rural settlements, that of the proportion of villages to hamlets. Village-to-hamlet ratios in Turkey of shows huge variations at provincial level (see Table 4.3). The primary reason for this is geographical—mountainous areas tend to have a relatively high numbers of hamlets compared to the number of villages. Thus, for example, the village-to-hamlet ratio in the mountainous Black Sea province of Ordu, provincial setting for the Çavdar-Mesudiye village-town project, is calculated at 1:6, and in Hakkari, another mountainous province, it is 1:5. In flatter provinces, there is a more even split, such as Diyarbakır’s 2:3 ratio (approximately), and Table 4.3 Number of villages and hamlets per province
(Source: Doğanay 1993, pp. 6–7) Province
Villages
Adana Adiyaman Afyon Ağri Amasya Ankara Antalya Artvin Aydin Balikesir Bilecik Bingöl Bitlis Bolu Burdur Bursa Çanakkale Çankiri Çorum Denizli Diyarbakır Edirne Elaziğ Erzincan Erzurum Eski ehir
720 347 428 559 346 1,075 585 475 485 911 246 329 298 797 189 697 572 471 744 406 741 255 558 561 1033 394
Number of Rural Settlements Hamlets Total 905 647 150 371 217 241 1,425 1,798 334 243 39 774 357 988 108 97 96 359 402 313 1,254 0 646 245 488 73
1,625 994 578 930 563 1,316 2,010 2,073 819 1,154 285 1,103 655 1,785 297 794 688 830 1,146 719 1,995 255 1,204 806 1,521 467
the integrated settlement network
133
Table 4.3 (cont.) Province
Villages
Gaziantep Giresün Gümü hane Hakkari Hatay Isparta Içel İstanbul İzmir Kars Kastamonu Kayseri Kirklareli Kir ehir Kocaeli Konya Kütahya Malatya Manisa Mara Mardin Mugla Mu Nev ehir Niğde Ordu Rize Sakarya Samsun Siirt Sinop Sivas Tekirdag Tokat Trabzon Tunceli Urfa U ak Van Yozgat Zonguldak Total
600 511 331 104 371 179 524 230 646 778 1,068 448 176 242 235 842 599 511 770 479 513 401 369 143 290 496 315 479 868 750 425 1,267 268 651 552 416 755 252 570 1,267 762 35,023
Number of Rural Settlements Hamlets Total 276 2,166 525 503 394 133 662 64 325 178 2,552 109 22 42 388 328 105 1,207 527 968 331 519 216 10 48 2,673 1,325 277 2,127 320 1,465 717 3 393 4,075 804 1,776 264 466 717 1,927 44,319
876 2,677 856 607 765 312 1,186 294 971 956 3,620 557 198 284 623 1,170 704 1,718 1,297 1,447 844 920 585 153 338 3,169 1,640 756 2,995 1,570 1,890 1,984 271 1,044 4,627 1,220 2,531 516 1,036 1,986 2,689 79,342
134
chapter four
in Van, home to the Dorutay-Özalp village-town project, where the ratio is about 5:4 (notwithstanding the low population density there, as in Hakkari). In the flattest, central and western, provinces ‘inverse’ ratios predominate, such as İzmir, which averages about two villages to each hamlet, and Bursa, at 7:1—to the extremes of Thrace/Trakiya, the Northwestern (European) part of Turkey, where Tekirdağ has a reported ratio of almost 90:1 and Edirne apparently has 255 villages and not a single hamlet! The third dimension of the high level of dispersion of small (rural) settlements goes down a level further, to an intra-settlement level: the distribution of houses within a settlement is irregular. Settlements tend to lack a center and the distribution of houses appears random. This observation is not new. The sociologist Nusret Kemal Köymen remarked in the 1930s that internal cohesion is absent in the small rural settlements (Köymen 1939b). The Hungarian geographer and ethnographer Béla Horvath, who traveled in central Anatolia (Ankara, Konya, Niğde, Aksaray, Nev ehir and Kır ehir) in 1913, noted that Turkish villages were marked by an irregular distribution of houses (in contrast to Tatar muhacir villages, which were characterized by plastered houses grouped around a village square and Circassian mühacir villages, which were characterized by a ribbon construction alongside a road) (Horvath 1997: 8). The settlement scatter was regarded as a collection of self-contained spaces, far away from the ideal of the nation as an economically, culturally and politically integrated space. Since the modernization project in Turkey was an elitist and centralist project, control over the rural periphery was considered to be vital. In order for the modernization project to be carried forward, it needed to go to the level of villages and villagers. This made the down-scaling of administration of crucial importance. The Administration Issue Descriptions of the administrative system in Turkey are characterized by such key phrases as ‘strong state’, ‘centralized bureaucracy’, ‘hierarchical structures’, and ‘military tutelage’ (Heper 1992). Without doubt, Turkey’s administrative system is centralist and military-guided, but whether Turkey can accurately be described as a strong state is far from indubitable, particularly at the level of the small rural grid.
the integrated settlement network
135
The Multi-Layered Administrative System The Turkish administration is a multi-layered system with at the top, the head of government and the head of state, respectively the prime minister and the president. The maximum period of time a prime minister, as head of the government, can remain in office until elections are called is five years. Historically, however, the reality of changing coalitions and fresh elections has led to a steady succession of cabinets at short intervals—59 in the 80-year period 1923–2003 (see Table 4.4). The head of state is elected by parliament for a period of seven years, and has a more stable history of continuous office (see Table 4.5). Table 4.4 List of Governments 1. İnönü Government (29.10.1923–6.3.1924) Prime Minister: İsmet Pa a (İnönü)
7. İnönü Government (4.5.1931–1.3.1935) Prime Minister: İsmet Pa a (İnönü)
13. Saraçoğlu Government (9.7.1942–9.3.1943) Prime Minister: ükrü Saraçoğlu
2. İnönü Government (6.3.1924–22.11.1924) Prime Minister: İsmet Pa a (İnönü)
8. İnönü Government (1.3.1935–1.11.1937) Prime Minister: İsmet Pa a (İnönü)
14. Saraçoğlu Government (9.3.1943–7.8.1946) Prime Minister: ükrü Saraçoğlu
3. Okyar Government (22.11.1924–3.3.1925) Prime Minister and Müdafaii Milliye Vekili: Ali Fethi Bey (Okyar)
9. Bayar Government (1.11.1937–11.11.1938) Prime Minister: Mahmut Celal Bayar
15. Peker Government (7.8.1946–10.9.1947) Prime Minister: Recep Peker
4. İnönü Government (3.3.1925–1.11.1927) Prime Minister: İsmet Pa a (İnönü)
10. Bayar Government (11.11.1938–25.1.1939) Prime Minister: Mahmut Celal Bayar
16. Saka Government (10.9.1947–10.6.1948) Prime Minister: Hasan Saka
5. İnönü Government (1.11.1927–27.9.1930) Prime Minister: İsmet Pa a (İnönü)
11. Saydam Government (25.1.1939–3.4.1939) Prime Minister: Refik Saydam
17. Saka Government (10.6.1948–16.1.1949) Prime Minister: Hasan Saka
6. İnönü Government (27.9.1930–4.5.1931) Prime Minister: İsmet Pa a (İnönü)
12. Saydam Government (3.4.1939–9.7.1942) Prime Minister: Refik Saydam
18. Günaltay Government (16.1.1949–22.5.1950) Prime Minister: emsettin Günaltay
136
chapter four
Table 4.4 (cont.) 19. Menderes Government (22.5.1950–9.3.1951) Prime Minister: Adnan Menderes
25. Gürsel Government (5.1.1961–20.11.1961) Prime Minister: Cemal Gürsel
31. Demirel Government (3.11.1969–6.3.1970) Prime Minister: Süleyman Demirel
20. Menderes Government (9.3.1951–17.5.1954) Prime Minister: Adnan Menderes
26. İnönü Government (20.11.1961–25.6.1962) CHP-AP Koalisyonu Prime Minister: İsmet İnönü
32. Demirel Government (6.3.1970–26.3.1971) Prime Minister: Süleyman Demirel
21. Menderes Government (17.5.1954–9.12.1955) Prime Minister: Adnan Menderes
27. İnönü Government (25.6.1962–25.12.1963) CHP-Yeni Türkiye Partisi-Cumhuriyetçi Köylü Millet Partisi-Bağımsızlar Koalisyonu Prime Minister: İsmet İnönü
33. Erim Government (26.3.1971–11.12.1971) Prime Minister: Nihat Erim
22. Menderes Government (9.12.1955–25.11.1957) Prime Minister: Adnan Menderes
28. İnönü Government (25.12.1963–20.2.1965) CHP-Bağımsızlar Koalisyonu Prime Minister: İsmet İnönü
34. Erim Government (11.12.1971–22.5.1972) Prime Minister: Nihat Erim
23. Menderes Government (25.11.1957–27.5.1960) Prime Minister: Adnan Menderes
29. Ürgüplü Government (20.2.1965–27.10.1965) AP-CKMP-MP-YTP Koalisyonu Prime Minister: Suat Hayri Ürgüplü
35. Melen Government (22.5.1972–15.4.1973) Prime Minister: Ferit Melen
24. Gürsel Government (30.5.1960–5.1.1961) Devlet Ba kanı, Prime Minister and Milli Mudafaa Vekili: Cemal Gürsel
30. Demirel Government (27.10.1965–3.11.1969) Prime Minister: Süleyman Demirel
36. Talu Government (15.4.1973–26.1.1974) Prime Minister: Naim Talu
the integrated settlement network
137
Table 4.4 (cont.) 37. Ecevit Government (26.1.1974–17.11.1974) Prime Minister: Bülent Ecevit
45. Özal Government (13.12.1983–21.12.1987) Prime Minister: Turgut Özal)
53. Yılmaz Government (7.3.1996–6.6.1996) Prime Minister: Mesut Yılmaz
38. Irmak Government (17.11.1974–31.3.1975) Prime Minister: Sadi Irmak
46. Özal Government (21.12.1987–9.11.1989) Prime Minister: Turgut Özal
54. Erbakan Government (28.6.1996–18.6.1997) Prime Minister: Necmettin Erbakan
39. Demirel Government (31.3.1975–21.6.1977) Prime Minister: Süleyman Demirel
47. Akbulut Government (9.11.1989–23.6.1991) Prime Minister: Yıldırım Akbulut
55. Yılmaz Government (30.6.1997–11.1.1999 ) Prime Minister: Mesut Yılmaz
40. Ecevit Government (21.6.1977–21.7.1977) Prime Minister: Bülent Ecevit
48. Yılmaz Government (23.6.1991–20.11.1991) Prime Minister: Mesut Yılmaz
56. Ecevit Government (11.1.1999–28.5.1999) Prime Minister: Bülent Ecevit
41. Demirel Government (21.7.1977–5.1.1978) Prime Minister: Süleyman Demirel
49. Demirel Government (20.11.1991–25.6.1993) Prime Minister: Süleyman Demirel
57. Ecevit Government (28.5.1999–18.11.2002) Prime Minister: Bülent Ecevit
42. Ecevit Government (5.1.1978–12.11.1979) Prime Minister: Bülent Ecevit
50. Çiller Government (25.6.1993–20.9.1995) Prime Minister: Tansu Çiller
58. Gül Government (18.11.2002–14.3.2003) Prime Minister: Abdullah Gül
43. Demirel Government (12.11.1979–12.9.1980) Prime Minister: Süleyman Demirel
51. Çiller Government (6.10.1995–16.10.1995) Prime Minister: Tansu Çiller)
59. Erdoğan Government (14.3.2003–) Prime Minister: Recep Tayyip Erdoğan
44. Ulusu Government (20.9.1980–13.12.1983) Prime Minister: Bülent Ulusu
52. Çiller Government (31.10.1995–25.12.1995) Prime Minister: Tansu Çiller
138
chapter four Table 4.5 List of Presidents
President
Term in Office
M. Kemal (Atatürk) İsmet İnönü Celal Bayar Cemal Gürsel Cevdet Sunay Fahri Korutürk Kenan Evren Turgut Özal Süleyman Demirel Ahmet Necdet Sezer
29.10.1923–10.11.1938 11.11.1938–22.05.1950 22.05.1950–27.05.1960 (ousted by military coup) 27.05.1960–28.03.1966 28.03.1966–28.03.1973 06.04.1973–06.04.1980 09.11.1982–09.11.1989 09.11.1989–17.04.1993 16.05.1993–16.05.2000 16.05.2000–
Turkey is subdivided into seven geographic regions, following the recommendations of the First Geographic Congress held in Ankara in 1941 (see Graph 4.2, Table 4.6). Four of the seven regions were designated as ‘coastal’ (kenar) regions because of their location adjacent to different seas in the north, north-west, west and south of the country. These regions are respectively the Black Sea region, the Marmara region, the Aegean region, and the Mediterranean region. Three of the seven regions were referred to as ‘Anatolian’, designating the unity of a large entity. These are Central Anatolia, East Anatolia, and Southeast Anatolia. The idea of regions as territorial entities is used in planning, in particular for the preparation of regional development plans by the DPT. Regions have not been given a clear administrative status, however, for fear of fostering regionalism. Regional administrative bodies do exist, such as GAP (specifically, the Regional Development Administration of the Southeast Anatolia Project), but their remit and area of jurisdiction is limited to issues of development and infrastructure. An exception was the State of Emergency Region, OHAL, a region that came into existence as a ‘state of exception’ on a constitutional ruling (notwithstanding the fact that it had the character of permanency). On July 19, 1987, the State of Emergency replaced martial law in the provinces of Bingöl, Diyarbakir, Elazig, Hakkari, Mardin, Siirt, Tunceli and Van, and the provinces of Adiyaman, Bitlis and Mus were defined as ‘neighboring provinces’. Later on, Adiyaman, Bitlis and the newly established provinces Batman and ırnak were also to be placed in OHAL, under, that is, the OHAL administrative authority. From the thirteen provinces covered by OHAL at its height, the four provinces
Graph 4.2 Regions in Turkey
the integrated settlement network
139
Ağrı Bingöl Bitlis Elazığ Erzincan Erzurum Hakkari
Kars Malatya Mu Tunceli Van Ardahan Iğdır
Osmaniye
(163,000 km2; 21% of total surface area)
(120,180 km2; 11% of total surface area)
Adana Antalya Burdur Hatay Isparta Içel Mara
East Anatolian Region
Mediterranean Region
U ak
Afyon Aydın Denizli İzmir Kütahya Manisa Muğla
(79,000 km2; 11% of total surface area)
Aegean Region
ırnak Kilis
Adıyaman Diyarbakır Gaziantep Mardin Siirt Urfa Batman
(75,000 km2; 9.7% of total surface area)
South-East Anatolian Region
Central Anatolian Region
Niğde Sivas Yozgat Aksaray Karaman Kırıkkale
Ankara Çankırı Eski ehir Kayseri Kır ehir Konya Nev ehir
(151,000 km2; 19% of total surface area)
Table 4.6 Regions in Turkey
Ordu Rize Samsun Sinop Tokat Trabzon Zonguldak Bartın Karabük
Amasya Artvin Bolu Çorum Giresun Gümü hane Kastamonu
(141,000 km2; 18% of total surface area)
Black Sea Region
Kocaeli Sakarya Tekirdağ Yalova
Balıkesir Bilecik Bursa Çanakkale Edirne İstanbul Kırklareli
(67,000 km2; 8.5% of total surface area)
Marmara Region
140 chapter four
the integrated settlement network
141
that remained when the regional authority was terminated in 2002 were Diyarbakir, ırnak, Hakkari and Tunceli.7 The OHAL region was run by a ‘super-governor’. Six men in all filled this office, given special powers above and beyond those of the regular provincial governors. Appointed by the president, the governor of the OHAL region was given sweeping powers, with two decrees in particular providing a legal framework by which to execute evacuation, granting, in effect a constitutional legitimacy to the whole resettlement program. Article 2 of Decree 285, issued with the establishment of the office of Regional Governor in 1987, gave the OHAL governor legal provision to evacuate and relocate whole settlements. The governor had the power to ‘order the temporary or permanent evacuation of villages, winter stations (for livestock) flocks and arable fields in areas within his territorial jurisdiction to make necessary arrangements for the general security and (. . .) order the resettlement or unification of inhabitants of such places.’ Further powers were granted with Decree No. 430, introduced on December 15, 1990, ‘Regarding Additional Measures in the State of Emergency Governor’s Office and during the State of Emergency Period’. Article 1 of this decree allowed the OHAL governor to exile people from the region. (IDP Project 2002). The OHAL governor was also given authority over the provincial governors in the region, including authority to coordinate action between them. The most powerful administrative sub-units in Turkey are the province (il) and the district (ilçe), which are administered by a governor (vali) and district officer (kaymakam) respectively. In the year 2000, the total number of provinces was 81 while the number of districts was 850 (DIE 2000). The Kurdistan part of Turkey comprises 19 provinces and 141 districts8 (Table 4.7). Each province (il) is administered by a governor (vali), who represents both the state (headed by the president) and the government (headed by the prime minister). The dual representation is reflected in the appointment procedure, with the governor nominated by the minister of the interior and appointed by the president. As chief executive of the province, the governor is responsible for the supervision of government
See also Chapter 2, note 36. The area covered by the master plan discussed in the previous chapter covers 12 provinces and 84 districts. It does not cover de provinces Ağri, Erzincan, Antep, Iğdır, Kars, Malatya and Urfa because the magnitude of village evictions in these provinces is considered of a lesser size, not needing a special plan for organizing return. 7 8
142
chapter four
Table 4.7 Number of districts in the provinces of the Kurdistan area in Turkey Province (and Provincial Town) Ağri Batman Bingöl Bitlis Diyarbakır Elazig Erzincan Antep Hakkari Iğdır Kars Malatya Mardin Mu Siirt ırnak Tunceli Urfa Van Total
Number of Districts (and District Towns) 7 5 7 7 10 9 8 9 3 3 7 13 9 5 6 6 6 10 11 141
and state functions, including the maintenance of security. The constitution stipulates that the central administration oversees elected local councils, and the governor oversees the elected councils in his jurisdiction, both the provincial council and also the municipal councils (so elected councils are overseen by both central administration and also the provincial governor). Such supervision may be far-reaching, for example with municipal councils having to submit their agendas to the governor for approval. Furthermore, governors were granted the extraordinary state-of-emergency powers under the decrees referred to above in respect of the OHAL ‘super-governor’—that is, they could order the evacuation and resettlement of villages and their inhabitants. Each district (ilçe) in a province has its own administration headed by a district officer (kaymakam). The district officer is appointed by the minister of the interior. Each district officer is responsible to the governor, serving essentially as his or her agent in supervising and inspecting the activities of government in the district. Decisions made by the district officers have to be submitted for approval to the governor. Districts are
the integrated settlement network
143
divided into sub-districts (bucak), headed by sub-district directors (bucak müdür). The sub-district directors are appointed by the minister of the interior on the nomination of the governors. They are responsible for law enforcement in the villages, and assisted by officials in charge of rural security, land titles, vital statistics, schools, and postal, telephone, and telegraph services. Other than the appointed governor, district officer, and the sub-district officers, these administrative units host several elected bodies for which every five year elections are held, supervised by the governor or the district officer or both. At the provincial level these elected bodies are the provincial general assembly (il meclis) and provincial council (il encümen). The number of these bodies corresponds to the number of provinces. The authority of the provincial general assembly (il meclis) and provincial council (il encümen) is mainly limited to budgetary matters, infrastructure and public services. Settlement units with a population of 2,000 or more inhabitants, including provincial and district capitals, have an elected municipality (belediye), headed by a mayor (belediye ba kanı), of which there were 3,215 in the year 2000. The authority of the municipal authority and the mayor is limited to such issues as the budget, housing plans, reconstruction programs, tax rates, and municipal services. The village is the lowest administrative unit. Because the maximum population of a village is 2,000, thus ruling out the establishment of a municipality, village administration is organized differently. The Village Act of 1924 states that villages are to be administered by a village headman (muhtar), a village council (köy derneği), and an assembly (ihtiyar meclisi). The village council and the village assembly exist mainly on paper, and the office of village-headman is the most important of the administrative bodies at village level (Scott 1968: 11).9 The institution of a village headman was first established in İstanbul in 1827. At the level of neighborhoods, the village headman was made responsible for security, order, and tax collection (Ortayli 2000: 108). City and town neighborhoods still have a headman, but the role is of much greater importance in villages. The Village Act, adapted in 1924 and since then amended and revised several times, gave to the (village) headman wide-ranging roles
9 Everyone entitled to vote is a member of the village association, its main task being the election of the headman. The main task of the village council is to supervise and control the headman, and to set priorities in village matters.
144
chapter four
and responsibilities: for official paperwork originating in the village related to marriage, births and deaths, maintenance of the electoral register, informing the government of any illegal activities of villagers, and informing villagers about laws and regulations, including communicating conscription orders for military service, matters related to agricultural production and agricultural subsidies (destek parasi), as well as representing the locality (village) at the relevant office, for example in applications for infrastructural works and with such external institutions as banks or cooperatives, giving him the power to make contracts (Scott 1968: 14). The Village Act essentially empowered and required the headman to act for and in the village representing his electorate on the one hand and, on the other, in the official capacity of representative of the state, through the offices of the governor and district officer. In theory, this gives the office of the muhtar considerable power, but in practice such is not really the case, and for very different reasons. First, the muhtar has the right to settle disputes, but no coercive sanctions at his disposal. He may call in the state, but ‘ is likely to be foolish’ if he calls in state sanctions against ‘kinsman and neighbors among whom he has to live out his days’ (Stirling1965: 262). Second, the village headman is not always, and in fact in most cases definitely not, the most influential person in the village. This is because those who have authority and power in the village, the large landowners (ağa), are not eager to become headman. Such local power holders do not want to be directly subjected to the pressure of government officials, which the headman is. Since local power holders want control over the headman, however, he might nominate or support a younger brother or someone else who is under his direct control. The village headman thus does not simply represent the state, but is also typically an interface in a power struggle between the state and local prominent persons. Third, although the village headman is a liaison-officer between the village on the one hand and the district and provincial administration on the other, he is not so effective in representing villagers in the administration. Even though he is formally elected to represent villagers, the headman has no voice vis-à-vis the governor or the district officer. The relationship between the headman and the governor or district officer is rather asymmetrical, and until the 1970s it was not uncommon for the headman to receive standing orders from the governor and district officer. Fourth, villagers do not necessarily always take notice of the provisions of the Village Act, and tend often to just plain ignore it—a tradition with a
the integrated settlement network
145
veritable history: ‘When village custom already provides an accepted alternative method of dealing with a problem where, as almost invariably, the village has no interest in altering the “status quo”, no notice is taken of the village law. Certainly, no one I met in the villages knows much more about it than that it exists’ (Stirling 1965: 272). Settlements with a population of less than 150 are not villages in law, they do not have an administrative status, nor host representative bodies. According to Tütengil (1970), the law makes four different suggestions for the administration of these small rural settlements, viz. no application of the Village Act, administration of these small settlements according to local customs, administration of these small settlements by the headman of a nearby village (or mayor of a municipality), or, the abolition of small settlements and the resettlement of their population in new, larger settlements. In practice, many of these small rural settlements are administered according to local customs without interference of the state. The PKK was able to benefit from this administrative vacuum, and establish some kind of administration in these settlements. The Dual Administrative System Turkey’s armed forces are not only an important political actor, continuously intervening in politics, not only by military coups, overt (in 1960, 1971, and 1980) or covert (1997), or secret operations (ranging from death-squad killings to producing fake polls for the manipulation of election outcomes), but also through established institutions with administrative agency. This makes Turkey’s system of administration dual, referring to the existence of both civilian and military administrative bodies. At the top of the military administrative system (until 2004 when it was reformed as part of the measures enacted in Turkey’s move toward accession to the EU) was the State Security Council MGK (Milli Güvenlik Konseyi ). The MGK was established by the military after the coup of 1960. The members of the National Security Council MGK were the president, the commanders of the army, gendarmerie, air force, the prime minister, and interior, foreign and defense ministers. The MGK was chaired by the president, but the powerful function of secretary performed by a military figure. Formally the MGK was meant to advise the government on matters pertinent to national security, but the institution arrogated to itself decision-making powers. Through the
146
chapter four
MGK the military supervised the government and assigned itself the task of being the guardian of the state against internal threats (Türke 1975); (Zürcher 1998: 303).10 The governor has similar powers to those of the military in areas under martial law, and it is not clear whether we should interpret this as the formal supremacy of a civil institution over the application of military means, or the militarization of a civil institute. This flexibility between military and civil functions of the institution of the governor in conditions of state of emergency or martial law recalls the Ottoman practice in which the governor was also the commander-in-chief over troops mobilized in his province, and, at least in the early period of the empire, personally headed these troops on military campaigns. During the life of the OHAL region, the strong linkage between governorship and security forces was reflected in the curricula of the governors. Five out of the six OHAL governors had a background in the security forces, either as head of the national police forces, or as head of provincial police forces. Only the last ‘super-governor’ had a career in civil administration rather than in the security forces.11 The governor did not only assume military functions (under the enforcement of a state of emergency), but administrative tasks also were appropriated by the military. During the 1990s, there were several examples of high ranking officers (general or brigade commander) deciding on the evacuation of villages for reasons of security, determining freedom of movement within a province, deciding on whether and where new settlements would be constructed, even though these were all tasks within the formal authority of the governor, not the military 10 The 2004 reform changed the duties, functioning and composition of the MGK. It is now a consultative body, with a civilian majority and civilian Secretary General, without executive powers or unlimited access to civil institutions. 11 Havri Kozakçıoğlu (OHAL governor between 1987 and 1991) had been head of police in Istanbul and reportedly received a training at Scotland Yard in 1987, the year he took office (Çelik 1994); Necati Çetinkaya (1991–1992) was head of police in the provinces of Hatay and Giresun); Ünal Erkan (1992–1995), was head of police of Ankara and Istanbul); Necati Bilican (1995–1997) was director-general of the police; and Aydın Arslan (1997–1999) was head of police in the province of Diyarbakir. Gökhan Aydıner (1999–2002) was a career bureaucrat having held governorships in various provinces since 1980). The first three governors (Kozakçıoğlu, Çetinkaya and Erkan) became active in the political party, DYP, after resigning as governor. Kozakçıoğlu became Deputy Chairman of DYP in 2001, and Çetinkaya a DYP M.P. in 1995 (he also went on to become one of the founders of the AKP). Erkan was minister of state in the cabinet of Mesut Yılmaz in 1996.
the integrated settlement network
147
(Pamukoglu 2003). While high ranking officers would usurp the authority of the governor, captains and majors sometimes usurped the authority of the district officer. Interviewees from district towns in Diyarbakir told me that the district officer, officially the highest representative of the administration in a district, was in reality subordinate to the military. This power relationship between the highest civil authority in the district and the military was underlined by the fact that the military officer, when he attended meetings of the District Administrative Committee (İlçe Idare Kurulu), took the chair ahead of the district officer. The District Administrative Committee is responsible for coordination in the district between state institutions. Revealingly however, it was not the highest commanding officer in the district (a major, binba i ), who directed the meeting, but a captain ( yüzba ı). The major did not consider the district officer his equal and therefore sent an officer of lower rank to the meetings of the District Administrative Committee. At meetings, the district officer addressed the captain as ‘my commander’ (komutanım), implying subservience. The administrative role of the military was more pronounced in the period, especially in the provinces, of the evacuations and resettlement program than it is now. That is not to say that the dual system is no longer operative, however. The state within the state, or ‘deep state’ (derin devlet) continues to function, as noted in the EU reports on Turkey, which refer to the armed forces as continuing to ‘exercise influence through a series of informal channels’ (CEC 2004: 15), and regret ‘persistent shortcomings’ or ‘insufficient progress’ in several areas including ‘civil-military relations’ (Eurlings 2006: 6). Constructing Administration The construction of state administration—by which we mean the expansion of authority and the spatial refinement of administrative units and bodies—has been a constant matter of concern in Turkey. This is not an issue confined to the modern, Republican period. On the contrary, today’s efforts at expanding authority and refining administrative units build upon developments made throughout the Ottoman era. At least two developments are of importance: firstly, that of direct rule (and related to that, the making of new administrative divisions), and secondly, the recognition of the village as an administrative unit. A brief
148
chapter four
discussion, drawing mainly on the work of Colin Imber, Mehdi İlhan, and Martin van Bruinessen, will help illuminate the issue (Imber 2002; İlhan 1981; Bruinessen 1988b). Provinces as political entities (fixed territorial units with governors appointed by the sultan) did not exist in the Ottoman Empire before the last decades of the 14th century. In the early years, the Ottoman rulers divided their territory into appanages for their sons, other family members and their most important followers. Although the division of territory was not yet a system of provincial government, two of the main elements of the later system seemed to have been already in place. One was the delegation of military command, the other the grant of appanages which carried with them the obligation to perform military service. By the late 14th century the first two administrative provinces (vilayet) of the Ottoman Empire had come into being: to the west of the Dardanelles lay Rumelia (Rumeli ), comprising the land acquired in Europe (Thrace), and to the east lay Anatolia (Anadolu), comprising the land acquired in Asia Minor (western Anatolia). As a result of eastward expansion, a third province (the province of Rum) came into existence, with Amasya as its chief town. By the middle of the 15th century, with the annexation of Karaman, the number of provinces increased to four. A list dated 1527 shows eight provinces, with Egypt, Syria, Diyarbakir, and Kurdistan added to the first four. A list prepared in 1609 by Ayn Ali, mentions 21 provinces.12 Some of the new provinces were gains from expansion of the growing empire, while others were the product of administrative divisions. A pattern was thus developed: provinces initially came into being through external development, territorial acquisition, and then subsequently through the internal reorganization of this newly acquired territory. Moreover, the expansion of the empire did not always result from 12 Ayn Ali was scribe and former intendant of the Ottoman imperial registry. The list of provinces was part of his work entitled Laws of the Ottoman Dynasty, Comprising a Summary of the Contents of the Council Register, presented to the grand vizier, Kuyucu Murad Pasha. This work comprised a ‘detailed description and analysis of the administration of the Ottoman feudal timar system’. It was composed from research carried out by Ayn Ali into the law books (kanunname) of decrees issued by the Sultans, and three sets of registers housed in the imperial registry, most importantly, the mufassal, which contained tax and census data compiled in the cadastral surveys. The work of Ayn Ali marks the developing administrative sophistication of the empire, and more than that: ‘Through Ayn Ali’s pen, the Ottoman law books and registers (. . .) became archetypes of the created order’ (Howard 2001).
the integrated settlement network
149
military conquest. Rather, there was a tendency to draw neighboring territories into the empire through a system of vassalage: ruling dynasties were kept in place but obliged to pay an annual tribute and the provision of troops to the Sultan’s army. For administrative, tax and other purposes, the provinces were generally sub-divided (although there were exceptional cases in which a province was not subdivided, such as Yemen). The Ottoman-Turkish word sancak (or its Arabic equivalent liva) was the name used to refer to the sub-units or districts within a province. Command of the provinces was given to a provincial governor (beylerbeyi), and of the sancaks to a district governor (sancak beyi), both appointed by the state and regularly rotated (Imber 2002: 177–215; Barkey, 1994 #3: 36, 79). The provincial subdivision into sancaks was usually based around the town from with the sancak took its name. Initially, the most important factor in determining the pattern of sancaks was the existence of former lordships or principalities, and of areas where marcher lords had acquired territories for themselves or their followers. In the immediate aftermath of a conquest, an Ottoman sancak would often retain the boundaries of the pre-Ottoman lordship. Within a generation, the inhabitants and their descendants would often have lost their non-Ottoman identity, and with their assimilation an area that had been formerly an independent principality became a standard Ottoman sancak. However, not all provinces and sancaks completely lost their separate identity—the Ottoman rulers had considerable difficulties in suppressing the dynasty of Karaman, for example. On the eastern border of the Empire, the first Ottoman survey of the sancak Amid (Diyarbakir), made in 1518, designated a group of fief holders as Akkoyunlu, apparently clansman or appointees of the Akkoyunlu dynasty that had ruled the area until 1503. The Ottoman rulers sought to counterbalance the influence of these representatives of the old order, however, by also awarding fiefs in the newly acquired land to men from distant areas of the empire. The 1518 survey of Amid shows that the Akkoyunlu held fiefdoms of modest value while the highest valued fiefdom went to a man registered as hailing from Rumelia. Independent dynasties survived in some areas of the Anatolian and Arab provinces. These were where the incumbent lords were locally too powerful for the Sultan to remove, such as the Kurdish lords. The register of 1527 makes a distinction between ten directly governed sancaks and seventeen indirectly governed emirates. Over time, the Ottoman Empire expanded its direct administrative control, but this
150
chapter four
was neither a smooth linear development nor an irreversible process. Several Kurdish emirates were turned into directly governed sancaks, only to later regain their autonomy—which itself was not assured. By way of example of this type of administrative history, the settlement and lands of Qulp (Kulp), northeast of the city of Diyarbakır, was an indirectly ruled sancak in 1540, according to the registers, an autonomous emirate in 1568 and again an indirectly ruled sancak in 1623; while Çermik, northwest of Diyarbakir city, was a directly ruled sancak in 1518, an autonomous emirate in 1527, again a directly ruled sancak from 1540 onwards, and a then again a sancak under indirect rule in 1632 (Bruinessen 1988a: 16–17). Ayn Ali made a note on the formal status of these entities. In listing the sancaks in the province of Diyarbakir, he notes that (at the time, in 1609) Diyarbakır had ten Ottoman districts and eight districts of Kurdish lords. In these latter cases, when a lord died, the governorship did not go to an outsider, but to the deceased lord’s son. In other respects, however, they resembled normal Ottoman sancaks, in that the revenues were registered and allocated to fief holders, who went to war under their own lords. In addition, Ayn Ali noted that there were five sovereign sancaks in Diyarbakır, which were outside the system of provincial government. Ayn Ali recorded similar independent districts in the province of Çıldır in the northeast of the empire and in the province of Van, where the Khans of Bitlis ruled independently until the 19th century. Land within a sancak usually fell into three categories: privately owned lands, land owned by a trust, and, the largest category, land which was at the disposal of the ruler and which the sultan allocated as fiefs. The timar fief system was simple. The sultan allocated land comprising a certain number of villages to a cavalry man, who collected taxes to ensure his livelihood, and raised a retinue for war. The fiefs were categorized according to their size (Barkey 1994: 36). In general the categories timar, zeamet or has referred to value of the fief, respectively, fiefs with a value of less than 20,000 silver coins (akçe), between 20,000 and 100,000 silver coins, and more than 100,000 silver coins (Imber 2002: 193–4). The smallest fief, timar, was the most common, and lends its name to the system as a whole. The allocation of land as timar provided the sultan with a standing cavalry army. A cavalry man had to bring with him on campaign not only his horse, but also his arms and armor, tents, and one or more armed retainers. The level of obligation was dependent on the value of the income from his timar. In the course of time, the
the integrated settlement network
151
timar as a system for maintaining a standing army collapsed, as the timar was increasingly allocated not to a cavalryman but to representatives of great men, known as ‘baskets’ who collected taxes instead of, and for, him, without any military obligation. The result was a loss in the number of timar that produced warriors (Imber 2002: 208). In addition, it should be noted that the timar system was maintained by war, as better and larger fiefs were allocated for prowess at war. As such the system worked well when the empire was expanding, a situation which was to change in the mid-seventeenth century (Barkey 1994: 37). Provincial and sancak governors were also military commanders, and the formal quasi-military powers granted to the OHAL and other governors under the state of emergency martial law can be seen as part of a heritage from Ottoman times. The abstract, political term sancak derives from its concrete meaning of a flag or standard: in times of war the cavalrymen holding fiefs in the sancak would gather, physically, under this banner. The troops of each sancak would than assemble as an army and fight under the banner of the governor (beylerbeyi) of the province. In this way, the structure of command in the battlefield mirrored the hierarchy of the provincial government. The governor was the commander-in-chief, the district governor (sancak beyi) commanded all cavalryman who held timar or zeamet in his sancak, the zeamet acted as officers, and the timar holders headed their retinues of armed men. In the Ottoman Empire the administrative system was not so much concerned with villages but with villagers (Ortayli 2000; Yılmazçelik 1995). The village as an administrative unit, was mentioned for the first time in the Governor and District Officer Law (Vülat-ı İzam ve Mutasarrıfını Kiram) and District Officer and Director Duties Statute (Kaymakamlıkların ve Müdürlüklerin Vezaifini amil Talimatı), two laws adopted in 1848 (Özefe 2001a: 6). However, a definition of a village was not given nor was it explained how a village could be established. It was the 1857 Land Code (Arazi Kanunnamesi) which first legally defined a village—as a settlement including a summer pasture, forest reserved for firewood, and other immovable goods. According to the 1864 Province Statute (Vilayet Nizamnamesi), administrative units were subdivided into province (vilayet), district (sancak), town (kaza), and village (karye), he same basic structure as today’s province (il), district (ilçe), town (bucak), and village (köy).13 Village representation was organized to coincide with
13
Prior to the republic, the main language of the empire was Ottoman Turkish
152
chapter four
the community (cemaat) living in the village. In mixed villages—Muslim and Christian—each community had its own representation, unless a community included less than 20 houses. In 1913 villages lost their legal status again with the adoption of the ‘İdare-i Umumiye-i Vilayet Kanun-u Mavakattı’ (Ortayli 2000: 98–116) and were given it back again with the acceptance of the Village Law of 1924, but by that time the Ottoman Empire had ceased to exist and the Republic of Turkey had been established. Adopted in 1871, the Province General Administration Statute (İdare-i Umumiye-i Vilayet Nizamnamesi) established an administrative link between town and village, the nahiye. The nahiye was granted administrative authority, and may be considered the Ottoman precursor to the center-village. It was a unit comprising a cluster of settlements, which, together with dependent villages and farms, was supposed to have a minimum male population of 500. Tribes could also be granted the status of nahiye, for example the Kîkî (Kîkan) tribe in Diyarbakir, in which the tribal leader was given authority over fifty villages inhabited by the Kîkî/Kîkan tribes people (Yılmazçelik 1995: 169). The nahiye had the status of a sub-district and clustered a number of settlements. For example, Beyü ebab and surrounding settlements was granted the status of a nahiye in 1871, and the Ottoman authorities also gave the status of nahiye to the Hertu , a tribe in the region of Beytü ebab. This meant there were two administrations in the area, one headed by the Beytü ebab nahiye official, Ali Ağa, the other one headed by the leader of the Hertu tribe, aban Ağa (Ortayli 2000: 105).14 New Villages Refinement of the administration was not only a major concern of the Ottoman rulers, but has also been also a growing concern of politi(Osmanlıca), which was Latinized during the mid-nineteenth century. Ottoman Turkish had a large Persian (Farsi) base and Arabic influence, and was composed of strands known as ‘Eloquent’ (Fasih), Middle (Orta), and Common (Kaba). To the mass of ordinary people using the vernacular, Eloquent Turkish, the language of the elite, the state and its legislation, was unintelligible. As part of the democratization project of the republic, Modern Turkish was established mainly on the common form (although heavily influenced by the more elite forms, such that it has changed dramatically in developing into contemporary Turkish). 14 Beytü ebab, in today’s province of Sırnak, became the site of the republic’s first Kurdish uprising in 1924, less than a year after its establishment.
the integrated settlement network
153
cians and administrators in the republican era, particularly since the 1960s, when Turkey was faced with an increasing population flow from the countryside to the cities. The entry of the issue onto the political agenda is ascribed to the person of Mustafa Ok, a prominent member of the Republican People’s Party (CHP). Mustafa Ok represented the Manisa district in parliament for the CHP party during the 1960s and 1970s, occupying the post of the Minister of Village Affairs and Cooperatives (Köyi leri ve Kooperatifler Bakanlığı) in Bülent Ecevit’s first coalition government, between January and November 1974. Some years before, in 1962, Mustafa Ok had written an essay on the idea of concentrating the rural population into large settlements, for which he was awarded the Yunus Nadi Prize from the newspaper Cumhuriyet. In this essay, Ok proposed to abolish all hamlets and reduce the number of villages, thereby enabling the down-scaling of administration and culture (although by the time he became minister, however, his party had already embraced the idea of establishing village-towns, based on the idea of uniting settlements, rather than concentrating their populations). In the essay ‘About the Establishment of Villages’ (Köylerin kurulu konusu), Ok proposed the unification of Turkey’s 40,000 villages (apparently he did not count hamlets). His proposal was to bring informed specialists together and let them decide on the unification of villages into 6,000 villages or 10,000 modern villages, each of these villages being a market center, an industrial center and a cultural center for the local population.—no small task, Ok acknowledged, but, as he wrote, ‘great deeds remain undone without great courage’ (see Box 4.1). Box 4.1 Extract from Newspaper Article by Mustafa Ok, 1962 (on the abolition of all small hamlets and villages and the concentration of their population in large rural settlements) (Source: Archive of the newspaper Cumhuriyet) The Issue of Village Resettlement We have forty thousand villages: for historical, social or religious reasons, they are spread out all over Anatolia. During the time that Anatolia was in the grip of the furious Jelali [Celali ] insurrections, the villagers moved away from the main roads, and to defend themselves withdrew into the hills and mountains. Villages became smaller and smaller, which is a terrible form of settlement, and we almost disappeared from the land in Anatolia as a result. It could be termed a form of nomadism: 42 percent of our 71 provinces could be termed ‘nomadic’.
154
chapter four
When we talk about a solution for the education problems, we have to talk about 40,000 schools and teachers for these schools, and for healthcare, 40,000 health visitors and doctors, and the same number of health clinics. And when it comes to electricity and irrigation, the number is 40,000 again. Road construction is the most important of all [our needs], but even if we dedicate the whole budget to this it will not be enough. Whichever need we focus on as the most important to tackle first, we simply do not have the means to provide all these basic services for 40,000 villages (. . .) There is no room to be optimistic as long as the number 40,000 remains. There must be a way to tackle this problem, however. Let us bring academics and experts together in order to unify villages, let them decide if and where it is possible to join villages together. They can decide if three, six or ten villages could be unified. Future industrial centers ought to be considered, big cities born in mind. How the land might be used effectively should be checked through maps, how electricity will be provided, where the main water reserves are, all this should be calculated. The new and modern villages should be established by main roads, whether six thousand or ten thousand, however many there may be. These new and modern villages should have electricity and diesel engines to become small industrial centers, production cooperatives should be set up. Agriculture should meet with engine and water. In fact, villages should enter the market. They should have internal imports and exports. They should have a cultural center. At present, our land reform is only symbolic and our social justice system no better. This should no longer be the case. Our national wealth is not increasing as quickly as our population. Apparently, there are 6,000 villages within or around forests. The villagers have axes and goats—they are enemies of the forest! This is the reality. No villager would want to harm the forest, but if it is the case of survival they will choose to live. If we build business enterprises in or around the forest, we will no longer have to fear the goats and axes. In the winter, our villages go into hibernation just like ants, as if we only have six months in a year in this country compared to twelve everywhere else. Remember, moreover, that half of the time of those six months is night. We have both direct (open) and indirect (hidden) unemployment. We cannot even discuss democracy and traditionalism in these conditions. Villagers will remain reactionary as long as they only go to the cities on market day and politicians will remain as demagogues so long as they do not visit the villages between one election and the next. With a resettlement plan, we are not discovering something new, we have other examples at our disposal. Frederick the Great established Germany in this way. This is how the villages of federal Germany were unified. Israel has gone down the same path, and France, Italy and Spain too. Let us not forget, great deeds remain undone without great courage.
the integrated settlement network
155
This approach of Ok’s was not new. Decades before, in 1937, then Prime Minister ükrü Kaya had concluded that ‘the principal shortcoming of our villages is that they are dispersed and small. It is evident that civilization (. . .) does not go to these small places’ (Özefe 2001a: 24). No serious attempts, however, were made to alter the situation. The abolition of small settlements and concentration of the population sometimes is interpreted as part of anti-Kurdish policies in Turkey. Yet at the time, the idea of abolition and concentration was part of a shared vision of societal transformation among intellectuals from both Turkish and Kurdish backgrounds. The abolition of small settlements and the concentration of their populations in settlements of considerable size was regarded as part of the ‘progress’ of history. In one of the most important treatises on the Kurdish issue, ‘Problems of the East’ (Doğu’nun sorunları, first published in 1966), M. Emin Bozarslan considers the blessings of a unification of villages in quite idealistic terms: ‘The villagers will be landowners, working with pleasure, living in clean and healthy houses, literate, confident, not worrying about anything, looking at the future with confidence, and buying plentiful and nutritious foodstuffs. Sturdy and happy children are having a good time playing. They wear clean clothes and do not walk around barefoot anymore. ‘The village youth is busy learning to read and write in village classrooms. They are equipped with necessary information about village development and have lessons in which they apply their knowledge. They spend their free time in the library, reading books, newspapers and magazines, increasing their knowledge and culture. From now on, they have a pen instead of a knife in their hands, a book instead of a pistol, and modern tools for production instead of a mauser. Those raised at regional schools practice agriculture, animal husbandry, viticulture, and apiculture with modern production methods and earn millions, both for themselves and for society. Doctors and engineers graduated from university will provide services to the people, illness will be left behind, free medical healthcare will protect people’s life, and every person will be given the same attention and affection. Villages will be spick-and-span, with electricity, park, school, road, classroom, library, shopping center and market. Every village will call to mind the city. Electricity plants make a roaring sound and, established at rivers and brooks will provide the surroundings with light and energy, irrigation canals turn dry valleys green like emerald, tobacco, rice, cotton, cereals will all be national treasures. Village roads will make relatives come and go, modern bridges will be built over the rivers and streams, no more the people in the east will cross streams and rivers swimming. Everywhere factories and laborers working there with pleasure. Barren mountains will be covered with forests. Minerals and petrol will be extracted, all national capital will be
156
chapter four used for the benefit of the people. Nobody exploits anybody, the ağalık, a remnant from the middle ages, will be something of the past, no trace remains of eyhlık, people are free, and talk and discuss without reluctance or shame about anything. ‘Everywhere joy, everywhere life, no longer people will kill each other. Everyone is an associate of the other, for the benefit of society and in solidarity they have become brothers. This brotherhood is not the fraternity of religious orders and tribal fraternity, but national brotherhood. Everyday hundred, thousands, millions of newspapers will be distributed and even reach the remotest villages and everyone will read them. In each house in the East, there is radio broadcasting in its language [Kurdish], with modern toilet, water, bath, electricity and well being. In each village a doctor, midwife, and clinic will watch for free over the health of the people.’ (Bozarslan 2002: 226–7)
The kind of approach espoused by Bozarslan and like-minded intellectuals was characterized by a paradigm of modernization informed by the idea of a necessary abolition of small rural settlements and concentration of their populations in urban settlement units. In the ‘official’ ideology of the nation-state, however, a natural relation was assumed between modernization and the enhancement of a Turkish identity, a relation which did not entail the demise of small villages and hamlets. Small, traditional rural communities were in important ways the keepers of the Turkish identity, an issue discussed in series of publications of the sociologist İsmail Be ikçi (1991a, 1991b, 1991c). Meanwhile, Kurdish nationalists (implicitly) referred to the development of a Kurdish modernity. Village Unification The abolition of hamlets and small villages and the resettlement of their inhabitants in larger settlements, was the object of serious research. Several times the costs of such an enormous operation were calculated. In 1963, the year after Mustafa Ok wrote his prize-winning essay, the cost for the resettlement of the inhabitants of all villages and hamlets into 10,000 larger settlement units of 10,000 houses each costing 20,000TL per household was calculated at two trillion Turkish liras, equivalent to 120 billion US dollars (Geray 1999). Three years after the military coup of 1980—but still under the prime-ministership of retired admiral Bülent Ulusu, who had been appointed by junta-leader Kenan Evren—the Ministry of Village Affairs and Cooperatives drafted a Model Village Project (Örnek Köy Projesi) in which the costs were cal-
the integrated settlement network
157
culated for the abolition of all hamlets and the establishment of their populations in larger settlements. The resettlement of the inhabitants of five hamlets (‘sub-village’) into a new village of 100 houses was costed at 415 million TL (1.5 million USD) per village. The total cost of this project was estimated at 4.3 trillion TL (15.6 billion USD). The inhabitants of the hamlets were supposed to pay 2 million TL or 7,000 USD per household for the privilege of their compulsory relocation into a modern settlement, which would have amounted to 200 million TL per village. The authorities would make up the remaining 215 million TL. Almost comically, it was also acknowledged that low-income villagers could not be expected to be able to find and pay 7,000 US dollars (Korkut 1987: 2–3). Another calculation was made in 1987. The abolition of 52,000 hamlets and the concentration of the population into 10,400 villages (at an average of five hamlets per new village) was costed at about 4 billion TL (2 million US dollars) per village, totaling some 41.6 trillion TL (20 billion USD). This did not include the resources needed to create a new framework of economic activities for the resettled population (Korkut 1987: 3). Somewhat similarly, in 1993 it was claimed that the concentration of the population—into small cities of up to 100,000 people and medium sized cities of between 100,000 and 1,000,000 people—was vital in order to fight what was called the ‘social fire of anarchy and disorder’ that weakened the ‘power and authority of the state in the Southeast’ (Akın 1993: 39). Using a medical metaphor, it was argued that the problem in the region should not be regarded as a cancer to be cut away, but instead as a sick organ to be nursed back to health for the benefit of the ‘national functional structure’. The small rural settlements were the cancer cells spread through the national body, and the remedy proposed the development of small and medium sized cities, to be called ‘attraction centers’ (cazibe merkezleri) (Akın 1993: 39–49) Center-Villages The idea of establishing center-villages appeared on the political scene sometime in the 1960s. Basically, the idea behind the establishment of center-villages was to identify rural settlements that could be equipped with the necessary means to perform central functions for rural settlements in the immediate vicinity. The maximum population of a centervillage—including both the core settlement and surrounding dependent
158
chapter four
settlements—was determined at 10,000. In the classification system of settlements (village, town, city, metropolis), the center-village falls within the category of rural towns to be administered by a sub-district director and municipality. The center-village would be given a formal administrative status and full authority over the rural settlements within its borders (like the nahiye in Ottoman times). The center-village did not entail a concentration of the population, but a concentration of services. Such a concentration of services would relieve the state of the burden of having to establish services in every single settlement, as described by Mustafa Ok, and avoided the need for an expensive and complicated resettlement operation (Günaydin 2001; Güven 1974; Güven 1977; Tütengil 1975). The most important envisaged services were cultural (education) and administrative (local administration, cadastre). It was assumed that, over time, center-villages would attract migration and develop into towns (administered by a sub-district or district officer), and that as a consequence the ability of the authorities to supervise and control the countryside would improve. It was thought that the integration of rural settlements into the national grid would produce a shared socio-cultural framework (including shared language and cultural values), and was supposed to be accompanied by the assimilation of ‘subcultures’ into the ‘national culture’ (Korkut 1987; Tütengil 1975); (Doğanay 1993). The idea to establish center-villages in Turkey on a nationwide basis was proposed in the third Five Year Development Plan, for 1973–1977, prepared by the State Planning Organization (DPT 1973). The provinces and government departments subsequently drafted their own lists of center-villages. The Ministry of Education proposed the establishment of 6,059 center-villages, the provinces 5,007, the Ministry of Village Works 4,351, and the Ministry of Agriculture 4,063 (Özefe 2001a: 71). The government determined on the (low) number of 3,546, but did make its selection of center-villages on the basis of criteria and figures reported by the provinces and various departments—including population density, quality of road connections, and available public services such as agricultural cooperatives, health clinics, schools and telephone exchanges, along with social and economic compatibility between settlements, and availability of (potential) natural resources (Geray 1977: 56). The size of the center-villages envisaged ranged widely, from 815 to 21,847 people, and the number of settlements in a sub-region dependent on a center-village from 1 to 64. About 71% of center-villages were standing villages with a muhtar; with just a few
the integrated settlement network
159
(not even 0.5 %) having the status of a hamlet (Geray 1977: 56–7). In December 1983, the Council of Ministers again decided to set about the implementation of a center-village policy, and published a list of 4,319 proposed center-villages (Korkut, 1987).15 According to this list, the smallest center-village would have a population of a mere 35 inhabitants (as compared to 815 in 1973), and the biggest, 29,374 (21,847 in 1973). The average size of the proposed center-villages and their dependent settlements was 1,402 inhabitants, with about 75% (against 71% in 1973) of center-villages having a population of less than 2,000, the minimum population needed for the establishment of a municipality. Table 4.8 extends Table 4.3 with the addition of the center-village numbers, by province, according to this 1983 proposal. Table 4.8 Number of villages, hamlets and center-villages per province
(Source: Doğanay 1993, pp. 6–7) Number of Rural Settlements
Adana Adiyaman Afyon Ağri Amasya Ankara Antalya Artvin Aydin Balikesir Bilecik Bingöl Bitlis Bolu Burdur Bursa Çanakkale Çankiri Çorum Denizli Diyarbakır Edirne Elazlğ Erzincan
No. of Villages
No. of Hamlets
Total
720 347 428 559 346 1,075 585 475 485 911 246 329 298 797 189 697 572 471 744 406 741 255 558 561
905 647 150 371 217 241 1,425 1,798 334 243 39 774 357 988 108 97 96 359 402 313 1,254 0 646 245
1,625 994 578 930 563 1,316 2,010 2,073 819 1,154 285 1,103 655 1,785 297 794 688 830 1,146 719 1,995 255 1,204 806
Number of Central Villages
80 41 57 66 31 140 59 40 74 127 27 45 54 91 32 76 70 65 81 50 64 45 60 51
A further count of villages and hamlets was made in 1990, and the list of centervillages and dependent hamlets was modified in 1997. 15
160
chapter four
Table 4.8 (cont.) Number of Rural Settlements
Erzurum Eski ehir Gaziantep Giresün Gümü hane Hakkari Hatay Isparta Içel Istanbul Izmir Kars Kastamonu Kayseri Kirklareli Kir ehir Kocaeli Konya Kütahya Malatya Manisa Mara Mardin Mugla Mu Nev ehir Niğde Ordu Rize Sakarya Samsun Siirt Sinop Sivas Tekirdag Tokat Trabzon Tunceli Urfa U ak Van Yozgat Zonguldak Total
No. of Villages
No. of Hamlets
1,033 394 600 511 331 104 371 179 524 230 646 778 1,068 448 176 242 235 842 599 511 770 479 513 401 369 143 290 496 315 479 868 750 425 1,267 268 651 552 416 755 252 570 1,267 762 35,023
488 73 276 2,166 525 503 394 133 662 64 325 178 2,552 109 22 42 388 328 105 1207 527 968 331 519 216 10 48 2,673 1,325 277 2,127 320 1,465 717 3 393 4,075 804 1,776 264 466 717 1,927 44,319
Total 1,521 467 876 2,677 856 607 765 312 1,186 294 971 956 3,620 557 198 284 623 1,170 704 1,718 1,297 1,447 844 920 585 153 338 3,169 1,640 756 2,995 1,570 1,890 1,984 271 1,044 4,627 1,220 2,531 516 1,036 1,986 2,689 79342
Number of Central Villages
91 54 77 62 41 37 51 35 80 41 108 104 104 60 22 29 16 166 56 93 81 84 59 54 35 22 27 96 47 54 81 59 47 147 27 80 68 38 68 44 72 92 84 4,319
the integrated settlement network
161
If we group the center-villages per region and province, some 26 percent of the center-villages were planned in East and Southeast Anatolia (the Kurdistan region in Turkey) ), as shown in Table 4.9 which collates the information of Tables 4.3, 4.6 and 4.8. In table 4.10 are listed the ten provinces with the highest absolute number of center-village. The three provinces with the highest absolute number of center-village—Konya, Sivas, and Ankara—are located in central Turkey. Of the ten provinces with the highest absolute number of center-villages, five are located in central Turkey; two provinces are located in the Aegean coastal area, two in the East, and two in the Black Sea coastal areas. The ten provinces with the highest absolute number of centervillages vary considerably. If we take the average number of settlements per center-village, Izmir has the lowest score with 9.0 settlements per center-village, while Kastamonu has the highest score with 34.81, almost four times as many. If we take the average number of hamlets per village, Ankara has the lowest score with 0.22 hamlets per village, while Ordu has the highest score with 5.39, higher by a factor of 24. Of these ten provinces, three—Ankara, Balıkesir and Izmir—are among the most developed areas in Turkey, while one, Kars, is among the least developed (Kele 2002, 377). A brief consideration of Table 4.10 indicates that little is to be derived from looking at the (1983) proposed implementation of the centervillage concept by reviewing just the absolute numbers of (projected) center-villages (grouped by province). Few conclusions can be drawn from this information other than that there seems to be no particular pattern in the spread of center-villages across the country when viewed as a whole. Nor is there is any obvious relationship of center-village totals to the number of inhabitants. The three biggest provinces in this respect—İstanbul, Ankara and İzmir—obviously reflect urban (metropolitan) populations, a matter that has no direct bearing on the issue of center-villages. Equally, and more interestingly perhaps, none of the three provinces with the lowest populations—Bayburt, Tunceli and Yalova—is among the ten provinces with the fewest center-villages. There is a correlation between the provinces with the highest number of proposed center-villages and those with the largest geographical area. In fact, the three largest provinces in Turkey—Konya, Sivas and Ankara—are also the provinces with the most center-villages, and in
İzmir (108–646–325) Manisa (81–770–527) Aydin (74–485–334) Muğla (54–401–519) Denizli (50–406–313) Afyon (57–428–150) U ak (44–252–264) Kütahya (56–599–105) Balıkesir (127–911–243) Bilecik (27–246–39)
678 5,144 2,819
İstanbul (41–230–64) Edirne (45–255–0) Tekirdağ (27–268–3) Kırklareli (22–176–22) Sakarya (54–479–277) Çanakkale (70–572–96) Bursa (76–697–97) Kocaeli (16–235–388)
351 2,930 947
421 3,047 4,595
Adana (80–720–905) Hatay (51–371–394) Içel (80–524–662) Mara (84–479–968) Antalya (59–585–1,425) Isparta (35–179–133) Burdur (32–189–108)
Mediterranean
802 6,439 2,644
Ankara (140–1,075–241) Çankiri (65–471–359) Eski ehir (54–394–73) Konya (166–842–328) Yozgat (92–1,267–717) Sivas (147–1,267–717) Kayseri (60–448–109) Kır ehir (29–242–42) Nev ehir (22–143–10) Niğde (27–290–48)
Central Anatolia
Total Number of Center-Villages: 4,319 Total Number of Villages: 35,023 (including center-villages) Total Number of Hamlets: 44,319
Aegean
Marmara
326 3,052 6,932
Bolu (91–797–988) Kastamonu (104–1,068–2,552) Zonguldak (84–762–1,927) Sınop (47–425–1,465)
West Black Sea (I)
627 5,289 15,701
Samsun (81–868–2,127) Çorum (81–744–402) Tokat (80–651–393) Ordu (96–496–2,673) Giresun (62–511–2,166) Trabzon (68–552–4,075) Amasya (31–346–217) Rize (47–315–1,325) Gümü hane (41–331–525) Artvın (40–475–1,798)
Middle and East Black Sea (II) Erzurum (91–1,033–488) Kars (104–778–178) Ağri (66–559–371) Erzincan (51–561–245) Malatya (93–511–1,207) Elazig (60–558–646) Tunceli (38–416–804) Bingöl (45–329–774) Bitlis (54–298–357) Mu (35–369–216) Van (72–570–466) 709 5,982 5,752
East Anatolia
419 3,810 5,587
Diyarbakır (64–741–1,254) Gaziantep (77–600–276) Mardin (59–513–311) Siirt (59–750–820) Urfa (68–755–1,776) Adiyaman (41–347–647) Hakkari (37–104–503)
Southeast Anatolia
number of villages and hamlets according to a 1990 count (bracketed: # center-villages, # villages, # hamlets) (Source: Doğanay 1993)
Table 4.9 Number of center-villages by region and province according to the decision of the Council of Ministers in 1983, and
162 chapter four
the integrated settlement network
163
Table 4.10 Provinces with the highest absolute number of center-villages according to
a 1990 count (Source: Doğanay 1993) Province
Number of CenterVillages
Number of Villages
Number of Hamlets
Total Number of Settlements
Average Number of Hamlets per Village
Average Number of Settlements per CenterVillage
Region
Konya Sivas Ankara Balıkesir Izmir Kars Kastamonu
166 147 140 127 108 104 104
842 1276 1075 911 646 778 1068
328 717 241 243 325 178 2552
1170 1984 1316 1154 971 956 3620
0.39 0.57 0.22 0.27 0.50 0.23 2.39
7.05 13.50 9.40 9.09 9.00 9.19 34.81
Ordu
96
496
2673
3169
5.39
33.29
Yozgat Erzurum
92 91
1267 1033
717 488
1986 1521
1.78 0.47
21.59 16.71
Middle Middle Middle Aegean Aegean East Black Sea I Black Sea II Middle East
the same order. This indicates that the provinces with the largest rural space tend to have a high number of center-villages.16 In order to use provincial statistical data in order to review the planning behind the center-village project, we need to look in more detail at the settlement structures of rural environments. If we take the center-village concept as a starting point, then we should first consider the statistical relationship between villages and center-villages. An approximate calculation of the average number of settlements per center-village is made by simply dividing the total number of settlements (79,342) by the total number of center-villages (4,319)—which yields the result of eighteen settlements for each center-village as an average. Table 4.9 lists the (twenty) provinces with an average of more than twenty settlements per center-village. The Black Sea coastal region predominates in this list, followed by the East and Southeast Anatolian regions. In general, these are the most mountainous provinces of Turkey, a geographical factor in settlement patterns noted above (mountainous areas tend to have the a
The figures for these provinces for population were—İstanbul: 7,195,773, Ankara: 3,236,378, and İzmir 2,694,770; Bayburt: 107,330 Tunceli: 133,584 Yalova: 135,121 (1990 DİE census figures); regarding geographical size, Konya covers an area of 38,157km², Sivas: 28,488km² and Ankara: 25,706km². 16
164
chapter four Table 4.11 Provinces with above average settlements per center-village
Province
Number of Settlements per Center-Village
Trabzon Artvin Giresun Sinop Kocaeli Urfa Samsun Rize Kastamonu Antalya Ordu Tunceli Zonguldak Diyarbakır Siirt Bingöl Adiyaman Yozgat Adana Elazig
68.0 51.8 43,2 40.2 38.9 37.2 37.0 34.9 34.8 34.1 33.0 32.1 32.0 31.2 26.6 24.5 24.2 21.6 20.2 20.1
Region Black Sea East Black Sea Black Sea Marmara Southeast Black Sea Black Sea Black Sea Mediterranean Black Sea East Black Sea Southeast Southeast East Southeast Middle Mediterranean East
higher number of hamlets). It would seem that the same topographical features that made the countryside such an effective theater of war for the PKK, the inaccessible mountain terrain, also led to a focus on the war zone in the center-village plan. State security analysts would draw obvious conclusions from this, as evidencing the argued necessity of a project of national integration at the administrative level in order to inculcate the values of the nation-state—the old problem, that is, of the perceived need to Turkify the territory. The extent to which such a conclusion would be warranted is questionable, however. A counter argument would assert that the coincidence between dispersed settlement patterns and a hospitable environment for non- or anti-national sentiment is not particularly indicative of a causal relationship between the two—rather, they both have themselves a common ‘cause’, which is the land, i.e. its mountains. It is instructive to compare the list of provinces with above average settlement-to-center-village ratios to that of the provinces with
the integrated settlement network
165
the highest absolute number of hamlets. Table 10 shows that seven of the provinces with the most hamlets are located in the Black Sea regions, with two in the Kurdish Southeast and one in the Mediterranean region. Table 4.12 Provinces with the highest absolute number of hamlets Province
No. of hamlets
Trabzon Ordu Kastamonu Giresun Samsun Diyarbakır Zonguldak Bolu Urfa Adana
4075 2673 2552 2166 2127 1995 1927 1785 1776 1625
Region Black Sea Black Sea Black Sea Black Sea Black Sea Southeast Black Sea Black Sea Southeast Mediterranean
When we compare the ten provinces with the highest absolute number of hamlets (Table 4.12) with the ten provinces with the highest absolute number of center-villages (Table 4.10), only two provinces match: Kastamonu and Ordu. This suggests a high absolute number of hamlets does not correspond per se to a high number of center-villages. However, when we compare the provinces with the highest absolute number of hamlets (Table 4.12), to those with the highest relative number of settlements (as compared to the number of center-villages) (Table 4.11) we find a high correlation: nine of the ten provinces with the most hamlets are also among the twenty provinces with a high ratio of settlements to center-villages (the exception being Bolu).17 This suggests that a high number of hamlets corresponds with a high relative number of settlements per center-village, a result that is fully compatible with the center-village idea of abolishing hamlets and concentrated settlement. From that perspective, a high number of
17 Five of the provinces with the most hamlets are among the top ten of provinces with a high ratio of settlements to center-villages (Trabzon, Kastamonu, Giresun, Samsun and Urfa), with a further three occupying positions in the top fifteen (Ordu, Zonguldak and Diyarbakır). Of these eight provinces, six are in the Black Sea regions, and two, Urfa and Diyarbakır in the (Kurdish) Southeast.
166
chapter four
hamlets does not correspond per se to a high number of center-villages, as that would merely reproduce a large number of settlements in areas with a large number of sub-villages. Instead, a high number of hamlets corresponds to a high number of settlements per center-village, which implies clustering. This is exactly the idea underlying the center-village concept. Although planning was in the spirit of the idea underlying the centervillage concept, however, its implementation was not. According to a study made of the implementation of the 1983 center-village policy (Doğanay1993), the project failed, because government investments were not in accordance with its center-village policy. If the centervillage policy had been properly implemented, public services would have increased in those center-village settlements. This was not the case, as is borne out by the overall figures. Whereas the establishment of schools in rural areas as a whole increased by 5.75%, for example, the establishment of schools in center-villages increased by a paltry 0.83%; and spending on healthcare services and infrastructure in rural areas increased by 29.65%, while the increase for center-villages was rather lower at 22.35%. Instead of concentrating services in center-villages, the net result of government action was in fact to reproduce the existing scatter by distributing services over the many small rural settlements. Village-Towns In the previous chapter, the village-town concept was linked to the person of Bülent Ecevit, the leader of the People’s Republican Party, the CHP, in the 1970s and of the Party of the Democratic Left, the DSP, in the 1980s and 1990s. The village-town idea was not his dream alone, but also part of a collective vision for a new Turkey enjoyed by the older generations of the nationalist Left. From early Republican Village Institutes to post-PKK village-towns, the revolutionary yet romantic concept of a statewide vision for the countryside, effective throughout the medium of a single, all-embracing rural plan of action—which would, its adherents believed, turn (backward) villagers into Turks—remained an efficacious ideal, one whose spirit infused the history of the republic and Turkish nationalism (a subject discussed in the following chapter).18
18 The Village Institutes were an attempt made in Turkey between 1937 and the mid-1940s to transform the Turkish countryside by means of a revolution in the edu-
the integrated settlement network
167
The implementation of the idea of village-towns was totally dependent on Ecevit assuming power in government. Prior to his term at the turn of this millennium with the DSP, Ecevit had led the CHP to power three times during the 1970s, and managed to extend his political influence over the State Planning Organization, the DPT. By the time of the DPT’s fourth Five Year Plan, the concept of center-villages had been erased, and replaced with that of village-towns. The idea of establishing village-towns was first mentioned (briefly) by the CHP as a part of its political program in its 1969 election campaign document (CHP 1969), and later in more detail in the election campaign documents of 1973 and 1976 (CHP 1969; 1973; 1976). These documents envisaged villagers, as bearers of national values, creating their own ‘rural cities’ as it were, a merger of village and city. A total of 5,000 village-towns were deemed necessary (Bulguç 1976), each of them to consist of a number of villages and hamlets and with a minimum population of 2,000 (the lower limit for the establishment of municipality).19 Regarding initial construction, the army was expected to contribute to the establishment of the village-towns, requiring, as they would, large investments in infrastructure, in particular for roads and highways (as we have seen, for example in the budgeting for the pilot project in ÇavdarMesudiye, Ordu). It was thought that allotting parts of the construction work to the army would save costs by using a conscripted work force cational system. Village institutes were established to train teachers for village schools, and to educate the rural population in the Turkish language and teach them the ideals of the Kemalist revolution. There were many different expectations from these institutions as agents for change in rural Turkey including bringing about a modernization of social relations, and an end to poverty and ignorance among the peasants, along with the creation of peasant intellectuals, an increase in agricultural productivity, and the spread of the Kemalist revolution in the countryside. The historical experience of the Village Institutes was rather controversial. The Village Institute system became a major focus of political and ideological debate in Turkey, especially in the 1950s and the early 1960s. Asim Karaömerlioğlu gives a good introduction in the controversies and debates attached to the Village Institutes experiences and the social and intellectual climate in Turkey from about the mid-1930s to the mid-1940s. He writes: Most leftist-oriented Kemalists saw in the Village Institutes the embodiment of Kemalist populism at its highest point, whereas many right-wing politicians and intellectuals condemned the Village Institutes and made them the scapegoats for their political ambitions and anti-communist hysteria. On the other hand, some socialists such as Kemal Tahir, a famous Turkish novelist, criticized the Village Institutes as being fascistic institutions by which the Single Party regime aimed to spread its ideology. (Karaomerlioglu, 1998: 48). 19 The army was expected to contribute to the establishment of village-towns, whose construction required large investments in infrastructure, in particular for of roads and highways. It was thought that allotting parts of the construction work to the army would save costs by using its conscripted work force.
168
chapter four
(the young men doing their required national service). This, of course, did not transpire. Regarding longer term economics, the merger of the village and the city constituted by the village-town concept was expected to rest on the modernization of agriculture (embedded in land reform and a cooperative movement), and in particular on a more effective division of labor and the development of small agro-industries (leading to increased production and an increased production range, and thus to extended job opportunities) (see Graph 4.3). The employment created by small agro-industries would, it was supposed, slow down or even stop migration to the cities, a migration which drained the villages and troubled the cities. Modernization of the village would also modernize life-styles, facilitating cultural integration of villagers into the national community (Oğuz 1976; CHP 1973; Geray 1977; Günaydin 2001; Güven 1974; Güven 1977; Topuz 1978). In retrospect, Bülent Ecevit related the idea of the making of villagetowns to the ideas and practices of the Ottoman reformer Mithat Pa a, who, in 1865, had established what he had called a ‘city-village’ ( ehirköy). Located in the Balkans, this city-village was a new settlement type based on the horizontal integration of four rural settlements. Not only were the names village-town and city-town very similar, Ecevit argued, but also their design: both Mithat Pa a’s city-town and Ecevit’s village-town were intended to overcome the radical division between village and city, and to lead towards the industrialization of agriculture (Ecevit 1993: 9–11). The village-town concept was subject to fierce criticism, however, in three main areas. First, it was argued that no real administrative upscaling would be effected, since the administrative status of the villagetowns had not been clarified. No legal measures were taken to arrange the authority of village-town councils. These councils, therefore, had a legal status as associations in accordance with public law, but not a legal status in accordance with administrative law. For this reason, villagetown councils would not be able to act as administrative bodies, and village-towns could thus not be considered legally established administrative units (Özefe 2001b). Second, the village-town approach aimed at a rural-industrialization, but it was disputed whether industrialization and urbanization could be considered independent from each other. Opponents argued that the establishment of village-towns would slow down social change, envisaged as the transformation towards a modern society characterized by the coincidence of urbanization and industrialization (Geray 1977). This reflected a basic premise of modernization
bound by a division of labor)
Graph 4.3 Graphical representation of a village-town (with individual settlements symbolized as different but equal,
the integrated settlement network 169
170
chapter four
theory, that urbanization and industrialization were the two processes guiding societal transformation: the village-town concept was criticized for decoupling industrialization from urbanization. (Proponents of the village-town argued the reverse, that such a decoupling was essential for the development of alternatives: the urbanization of industry had drained villages and flooded cities, and the industrialization of the countryside had the potential to turn the tide.) Third, it was argued that the concept had been proclaimed in an extremely idealistic manner and that there were problems of effective implementation in a country such as Turkey. It was argued that the idea of village-towns would only work in areas where there was a dense scatter of settlements (such as in Israel, where the agricultural-economist Halperin had envisaged a similar model), but not in Turkey where settlements were geographically distant from each other (Güven 1977). The first village-towns were planned in 1978 in the provinces of Van and Urfa. In Van, the establishment of village-towns was linked to the development of animal husbandry, while in Urfa, the establishment of village-towns was associated with the development of irrigated agriculture (later administered by GAP). Van and Urfa are both part of the area called Kurdistan and mainly inhabited by non-Turkish populations. Van has a predominantly Kurdish population, while the population in Urfa is composed of two main groups, Kurds and Arabs. Both provinces also have international borders, Van with Iran and Urfa with Syria, and the populations on both sides of the borders maintain tribal and familial relations. These borderlands were frontier areas for the Turkish state, and the village-towns were supposed to integrate the target populations more firmly into the settlement space of Turkey, establishing closer socio-economic, cultural, and administrative links (channels) between village-town settlements and higher order centers—to which purposes the establishment of good road connections was given high priority.20
20 The construction of road connections in the area was developed as a separate project, the ‘Reach the Villagers Project’, KUP (Köylüye Ula ım Projesi). The development of employment was organized in a ‘Work for Villagers Project’, KİP (Köylüye İ Projesi), which primarily consisted of the employment of (female and young) villagers for the clearing of hundreds of hectares of land covered with stones and therefore unsuitable for agriculture (using cheap labor for heavy, unskilled work without a future) (Topuz 1978: 156).
the integrated settlement network
171
Agricultural Towns The idea of agricultural towns, put forward by Alparslan Türke and his Nationalist Action Party, the MHP, resembled that of villagetowns in many respects. Nationalists on the Right thought that the development of small-scale agro-industries and the mechanization of agriculture would reanimate village life, which was believed to be the basic component of national life. Again, it was thought that the establishment of agricultural towns would slow down rural-to-urban migration and obviate the need for people to make the giant leap from village to metropolis. The nationalist Right proposed the establishment of agricultural towns encompassing eight to twelve villages, with an attraction center (cazibe merkezi) which would be the growth point of the agricultural town, the central settlement within a cluster of settlements (Türke 1975; Yahyaoğlu 1971). Türke suggested the establishment of 4,000 such agricultural towns (like Mustafa Ok, Türke estimated the number of villages in Turkey at 40,000) (Türke 1975; Tütengil 1975). Both village-towns and agricultural towns were based on the idea of stimulating a simultaneous process of vertical integration, through the transformation of rural settlements into centers of a higher order, and the industrialization of rural entities—as advocated by Halperin and Friedmann. Conclusions This chapter discusses ideas and proposals that evolved during the latter part of the twentieth century for the reorganization of rural space in Turkey, in particular the concepts of village-unification, village-towns and center-villages. It is argued that these concepts, developed in the 1960s and 1970s, were aimed at taking modernization to the countryside by means of a reorganization of space, an approach that employed a focus on territory as the locus of modernization, rather than people, modernizing villages by (re)construction rather than villagers by having them migrate to cities, the conventional view on modernization. As such, this represented a proactive as opposed to laissez faire approach, and not only to modernization but also to the perceived problem of a disconnection between the urban centers and dispersed rural villages and hamlets. The existing loosely connected settlement grid was regarded as non-compatible with the idea of nation (along the horizontal axes)
172
chapter four
and the state (along the vertical axes), which were ideally characterized by high level of integration and uniformity. The chapter presents what is essentially a historical overview of the recent evolution of the rural reorganization approach—yet these ideas remain today, at the turn of the 21st century, key-concepts in official resettlement and return programs. It is in the light of this history in general, and in terms of the center-village and village-town projects in particular, that the resettlement of Kurds evicted from their homes and villages by the military has been considered by the civil authorities during and since the 1990s. Through the concepts of village-town and center-village, Turkey could reconstitute the Kurdish issue as a developmental issue—specifically, a lack of social, economic, cultural and administrative development in the region. Resettlement of the displaced became absorbed by a rural development paradigm. However, the paradigm of rural development, effected through the establishment of village-towns and/or center-villages, became itself a contested paradigm in the southeast of Turkey, mainly because of the rise of a security paradigm. Lack of funding, in particular the World Bank non-support for village-towns in the war zone, and the identification of this model with the politician Bulent Ecevit, further undermined the model’s viability. Whether the territorial approach to modernization outlined in this chapter will reemerge remains an open question. What is clear is that the return-to-village issue and, more generally, the ‘problem of the Southeast’, as it is sometimes termed, do not go away, and nor does the imperative of nation-building in a country that is, after all, still less than a century old, and with which the issue of Kurdish nationalism in Turkey is so intimately interwoven. Indeed, the war initiated by the PKK was in many ways defined by the history of Turkish nation-building, a subject to which we now turn.
CHAPTER FIVE
SETTLEMENT ISSUE AND SETTLEMENT ACTIVITIES The Colonization of Peoples and Territories Where is the nation? It is not there yet. Yakup Kadri Karaosmanoğlu Today the Turkish villager is about to lose his existential self. (. . .) There are brothers who have forgotten their language and talk another language. There are brothers who consider it an insult if you called them Turk. It is our responsibility to construct their villages and to make our brothers talk, dress, and live like us. Abdullah Ziya1
Introduction In the previous chapter a genealogy of key concepts brought us to approaches for the reorganization of rural space (new settlement types and structures). This had a history pre-dating the military coup of 1980, the subsequent growth of the PKK, its war with the Turkish state and the military response of wholesale evacuations and village destruction in large parts of the Southeast—which itself eventuated in the employment of the new settlement types and structures in civil authority plans for a rehabilitation of the region. A driving force behind the initial ideas for a reorganization of rural space was the desire for vertical integration (the development of an intermediate settlement connecting the rural grid to regional centers). In this chapter, the desire to develop a new settlement type and structure is related to ideas about nation-state building that actually emerged much earlier, in the 1930s. More specifically, this chapter will explain how the quest for nation-state building tools in the 1930s produced the idea of a new settlement type going beyond the modern rural-urban divide. This new settlement type, referred to as rurban, was not supposed to contribute to vertical integration (i.e. via urban centers, as was the case with the new settlement types developed in the 1960s and 1970s) but to completely erase the rural-urban divide. 1
Karaosmanoğlu 1934: 153 and Ziya 1933.
174
chapter five
The somewhat ad hoc use of the approach of rural reorganization in attempting to set about dealing with the problem of the displaced persons and denuded territory of the Southeast in the 1990s thus in fact had its origins in a far more radical approach to the organization of socio-economic space, geo-political territory and contemporary life. This chapter will commence with a discussion of the 1934 Settlement Act, Law Number 2510. In academic literature (particularly in Kurdish studies), Settlement Act No. 2510 has been generally regarded as a means for the colonization of Turkish Kurdistan and the assimilation of Kurds through forced and collective resettlement. This view is not necessarily wrong, but it is narrow and incomplete. The argument developed here is that Settlement Act, Law Number 2510, emerged against two different backgrounds: on the one hand, there were the policies and practices of deportation of rebellious Kurds, indeed, but on the other there was the settlement in Anatolia of Muslim populations from former Ottoman territories. The new law had the overall objective of creating a general framework for national settlement in Turkey. Although the view is not pushed here, it could be argued, somewhat against the conventional currents of thought operative in Kurdish studies, that the employment of the 1934 Settlement Act cannot be understood as an instrument to quell insurrection in the Kurdistan region, but must be analyzed as part of a larger, positive objective of creating a homeland of the Turks. In the second part of this chapter, it is argued that the idea was rising in the 1930s that nation-building, a major preoccupation of intellectuals and politicians in the nascent country, should include the design of a new village types, intended to somehow imbue their inhabitants with ‘Turkishness’. Nations come about through processes of active creation, as considered in the first chapter, and Anatolian villagers, it was believed, needed to be guided and organized so as to identify themselves as Turks, to incorporate the nation and embody its ideals, and thence become part of the body politic. This idea resulted in series of architectonical blueprints for ‘the Turkish village’—to which a sociological dimension was added by the prolific thinker Nusret Kemal Köymen. Köymen saw villages (i.e. small rural settlements) as the basic cells of society, yet undermined by two processes: the coincidence of urbanization and industrialization in western modernization (alluded to in the last chapter, in connection with the development, and criticism, of the villagetown concept), and the separation of culture from civilization (to be traced back to the division in Christian thought between the sacred
settlement issue and settlement activities
175
and the profane, the schism of the spiritual and the material—and in which, of course, the sacred spiritual was vanquished by material determinism and the industrial imperative of modernity). Köymen’s imagination of the ‘new village’ is the subject of investigation of the last part of this chapter. Settlement Act The Settlement Act No. 2510 may quite possibly be the most important piece of legislation for the organization of a spatial framework for national settlement. The Bill was passed by the Turkish Parliament on June 14, 1934, and reported in the Official Newspaper (Resmi Gazette) exactly one week later, upon which the law became effective. The Act was hailed by some at the time as providing the means for nation building in Anatolia (Köymen 1934), but denounced by others later as a tool for nation destroying (Besikçi 1991). It categorized Turkey’s inhabitants into three groups (Articles 12, 13, and 14) and divided Turkey into three zones (Articles 2, 12, 13, and 14). The three groups were: those who spoke Turkish and were considered to be of Turkish ethnicity (Turkish speaking inhabitants of Anatolia and Turkish speaking immigrants); those who did not speak Turkish but were considered to be of Turkish ethnicity (non-Turkish speaking ‘Turkish’ immigrants, and—presumably—Kurds, who were deemed Turks by the new regime); and those who did not speak Turkish and were considered not to be of Turkish ethnicity (Arabs and non-Muslim minorities, the two ‘others’ in the East and the West). The three zones were: those areas in which it was deemed desirable to increase the density of the culturally Turkish population (Zone 1); those areas in which it was deemed desirable to establish populations that had to be assimilated in the Turkish culture (Zone 2); and all those areas which it had been decided should be evacuated for economic, political, military, or public health reasons, and where settlement was forbidden (Zone 3). The law determined different settlement policies for different groups in different zones, and granted powers to the Minister of Internal Affairs to implement these policies. Immigrants of Turkish origin and culture were to be settled in Zone 1, predominantly areas along the international borders as well as particular parts of the Kurdish southeast. A detailed specification of a Zone 1 settlement area was made in decree number 2/12374 issued on 24–11–1939 (Box 5.1), as an area composed mainly
176
chapter five Box 5.1 Specification of Areas in the Southeast where it was Deemed
Desirable to Increase the Density of the Culturally Turkish Population (Zone 1) (Birinci iskan mıntıklarında toprak tevziatına dair olan talimatnamının kabulü hakkında kararname) (Source: Kökdemir 1952: 167–169) 1. A 20km strip alongside the railway starting in Sallar and passing or will pass Ergani, Diyarbakir, Bismil, Be iri, Kurtalan, Baykan, Bitlis, Tatvan, Gevas, Van, and Özalp, and the train track that starts in Kurtalan and runs to the border with Iraq. 2. A 20km strip alongside the Fırat Bridge and the Yolçatı-Elazığ—Diyarbakir railway. 3. A 20km strip alongside the Divriği, Kemaliye, Erzincan, Tercan, A kale railway, the railway from Çobandede to Sarıkamı , and the railway from Sarıkamı to Benliahmed. 4. A 20km strip alongside the Malatya, Fevzipa a, Divriği railway. 5. A 25km strip alongside the railway in Urfa and Mardin provinces. This railway indicates Turkey’s southern (international) border (with Syria). 6. A 25km strip alongside the international borders with Syria, Iraq, Iran and the Soviet Union. 7. The Diyarbakir—Mardin macadamized road, and the roads DiyarbakirSiverek-Urfa-Akçale, Urfa-Birecik, Diyarbakir-Silvan-Ziyaret, DiyarbakirHani-Garahin-Çapakçur, Bitlis-Mu -Bulanık-Malazgırt, Mu -Hınıs, Mu -Solhan, Van-Muradiye-Ho ab and Ba kale-Yüksekova. 8. The road from Pasinler to the Cobandede bridge and Velibaba, Ele kirt, Karaköse, Diyadin, Doğubeyazit; the roads Karaköse-Hamurderesi-Patnos, Doğubeyazit-Muradiye, Erzurum-Ta kesen-Hınıs, Çobandede-Sarıkamı -Benliahmet-Kars, Kars-Kötek-KarabulakDoğubeyazit-Kağızman and Tuzluca-Iğdır- engelgediği-KaraköseKağızman. 9. A strip of 15km alongside the macadamized roads: Elazığ-PetrekMameki-Seyyithan-Darboğaz-Pülümer-Mutti Bridge-Sanboğazı, Pertek Hozat-Karaoğlan-Mara al, Çakmak-Zeyni, Gediği-Erzincan, RefahiyeErzincan-Tercan, Erzincan-Gümü hane, Elazığ-Palu-Bingöl-Solhan, Palu-Karakoçan-Kığı-Tercan, Elaziğ-Yolçatı-Kömürhan Bridge, ElazıgGezin-Maden-Ergani and Elazığ-Keban-Arapkir. 10. The roads Malatya-Hekimhan-Kangal-Ula and Pazarcık-ElbistanDarende-Gürün-Ula ; Elbistan-Göksün is a number one zone alongside the Divriği-Zara-Koyulhisar road. A strip of 15km alongside Zone 1 at the Mara -Göksun-Pınarba ı road. 11. A circle of 20km around the iron mine in Divrik, the copper mine in Ergani, and the chrome mine in Guleman, and the processing plants in Malatya center and Lake Van. 12. A strip of 25km north and northwest of Lake Van, from Erci to Tatvan. 13. A strip of 25km alongside Murat River between Palu and Mu .
settlement issue and settlement activities
177
of land on either side of roads and railways in the Southeast, and a 25 kilometer wide strip along the borders with Syria, Iraq, Iran and the Soviet Union. Alongside these roads, railways, and borders (strategic locations, either linking the state center with a rebellious periphery or demarcating international borders) Turkish speakers and villagers of Turkish ancestry had to be ‘tied to land’ (article 2 of decree 2/12374). Non-Turkish speaking inhabitants of these areas (Kurds and Arabs mainly) were to be concentrated in settlements and were not allowed to occupy more than 20% of the land. Under the provisions of the law Kurds could be deported to Zone 2, predominantly areas in the western parts of Turkey (the fate of many Kurds from the Dersim area after a rebellion in that area was crushed in 1938). Zone 3 areas were mainly small rural areas in mountainous areas. These small rural settlements were not only considered to pose a significant security problem (supervision of the population was deemed impossible in the scatter of small settlements composed of just a few households located in areas difficult to access), but also regarded as incompatible with modernity. Following the Turkish sociologist İsmail Be ikçi (Be ikçi 1991a), the 1934 Settlement Act has been mainly interpreted and criticized as a law intended for the assimilation of Kurds by means of forced and collective resettlement (the law was used for the forced settlement of Kurds from Sason, Zeylan, Ağri and Dersim to western Turkey). Be ikçi would later advance the argument by positing the thesis of a colonization of Kurdistan (Be ikçi 1991b). Although, however, the assimilation of Kurds by means of resettlement was a major concern, the law was equally concerned with the settlement and assimilation of Muslim migrants from former Ottoman territories. This is illustrated by the patchwork of decrees and laws the Settlement Act synthesized, legislation that had been adopted over the previous two decades (the 1910s and 1920s) and regulated the settlement of Muslim migrants from lost territories as well as Kurdish deportees (Kökdemir 1952). These laws and decrees included, among others, two settlement acts. The two settlement acts—one passed on the May 31, 1926 (İskan Kanunu No. 885) and the other two months later, on August 1, 1926 (İskana Ait Muhtira)—were mainly concerned with spelling out the conditions for the settlement of immigrants from Greece, the Balkans, and the Soviet Union. According to these laws, those who were not considered cultural Turks (‘Türk harsına dahil olmayanlar’), i.e. anarchists, spies, gypsies, or people suffering from venereal diseases) were not entitled to settlement
178
chapter five
in Turkey (Kökdemir 1952: 25–7, 192–3), a provision which was included in the 1934 Settlement Act, No. 2510. Settlement Act No. 885 also mentions the abolishment of small dispersed villages and the concentration of their populations into appropriate centers (article 3), a provision which was elaborated on in decree No. 10822, arranging the resettlement of people living in mountainous areas, passed on March 18, 1933.2 So, not only should the 1934 Settlement Act be appreciated in its full context, applying to Muslim migrants from former Ottoman territories and not aimed solely at dissenting and rebellious Kurds, but also it should be recognized as sketching out the first plans for the rural reorganization program, half a century before the implementation of the first center-village projects. Resettlement Origins of the Settlement Act Some of the arrangements made in the settlement laws are a continuation of practices from the 1910s, the final years of the Ottoman period when nationalism had already occupied the political imagination. Ottoman authorities had tried to organize the settlement of immigrants from lost territories by establishing immigrant settlement commissions İskan-I Muhacirin Muhacirin ve A air Müdiriyeti Umumiyes (the İskan-I Muhacirin, later renamed Muhacirin ve A air Müdiriyeti Umumiyesi) which determined areas appropriate for settlement, and, if deemed desirable, made various grants to the Muslim immigrants, including land, livestock, seeds, and dispensation from taxes and military conscription (Doğanay 1996; Dündar 2001: 177–200).3 A dispersal of populations characterized the 2 Some laws and decrees were, of course, aimed at Kurds—two 1927 laws, for example, Laws No. 1097 and No. 1178, which were explicitly designed to deport Kurds, from Beyazit and Diyarbakir. Law 1097, the Law on the Deportation of Some Individuals from Eastern Zones to Provinces in the West (Bazı E hasin ark Mıntakasından Garp Vilayetlerine Nakline Dair Kanun), arranged for the deportation of 14,000 families from Beyazit to provinces in the West (Kökdemir 1952: 28–9); under the provisions of Law 1178, about 500 people from Diyarbakir who had knowledge of or were actively involved in (bil-fiil ) the eyh Said uprising were deported to İzmir, Aydın, Manisa, Antalya, and Bursa (Beysanoglu 2001b: 1026; Kökdemir 1952: 30–1). Law 1178 gives the Council of Ministers authority to the to implement the provisions of Law 1097 (1097 Numaralı Kanun Hükmünün fe’ine İcra Vekilleri Heyetinin Mezun Olduğuna dair Kanun). 3 The Turkish word muhacir is translated here as immigrant, to denote people coming from lost provinces into the shrinking (rump) Ottoman Empire, or from the ‘outlying’ lands of the old empire into the new nation-state of Turkey (essentially, that is, people
settlement issue and settlement activities
179
late Ottoman settlement politics. Albanian and Bosnian immigrants for example, were settled in a dispersed pattern across Middle and Eastern Anatolia, where they were not allowed to form more than 10% of the local population in any one area. Kurds who had fled from the East (a front with the Russian army) or who had been deported by Ottoman military authorities for security reasons were transported to Middle and Western Anatolia. The Kurdish migrant populations were broken up in groups not exceeding 300 persons and dispersed in such a way that they would not make up more than 5% of the population in the areas where they were resettled. Also, leaders were settled separated from their followers (Bruinessen 1997; Dündar 2001: 247). In short, resettlement practices became organized as a practice of assimilation, as part of the process of nation building.4 Population Movements and the Drawing of New Maps The Ottoman Empire, we would say with today’s framework of reference, was composed of many peoples—among them Turks, Kurds, Greeks, Arabs, Armenians, Bulgarians, Serbs and Albanians. Prior to the nineteenth century, however, peoples were little categorized by
coming into Anatolia), with migrants used to refer to those moving across it (mainly Kurds, under duress). The word muhacir was borrowed by Ottoman Turkish from the Arabic transliterated as hajara, and originating in the root hgr with a core meaning of ‘depart’; the word muhaci was replaced in the 1930s by the modern Turkish göçmen. The use of the single English terms immigrant and migrant here covers a wide range of reasons for the movements of the peoples concerned, which might be categorized as ranging between chosen and forced. Other words which could be used would include settlers and the resettled, people who were evacuated or expelled, evicted or transported, refugees and people who escaped or fled, émigrés and expatriates, people who were repatriated or exchanged, etc.—each of these nuanced terms specifying subtle distinctions in the particular socio-political situations and human motivations, but which are generally not employed here, for reasons of simplicity and brevity. 4 This organization of assimilation is highly unusual and very revealing of the Ottoman/ Turkish mindset. The usual practice in such cases, and, indeed, in general for all cases of immigration, does not involve a planned dispersal. Governments tend not to make special provisions for people entering the country to live, or they keep refugees in centers which have sufficient infrastructure and other resources to cope with the influx, or else near the border across which the refugees came, envisaging return (e.g. in recent years, Afghani refugees to Iran), or else, if incomers are dispersed, it is for other reasons (e.g. the current policy in Britain for asylum seekers, in which the aim of dispersal is to prevent any concentration of associated problems). A centrally planned assimilation policy of the kind enacted by the late Ottomans and early republicans is quite possibly unique in human history.
180
chapter five
ethnic origin (let alone their national identity), but by religion. Religious communities, although subjected to the Sultan, were largely left alone to govern their own affairs, for which they had their own (religious) institutions. In the Ottoman countryside, Muslim and Christian communities (of different persuasions, especially Sunni, Shi’ite and Alevi, and Orthodox and Catholic) coexisted in neighboring settlements or even lived together, sharing villages—and they certainly shared towns and cities, along with other religious groups, notably Jews. Nevertheless, the composite character of the Empire found itself under siege from the emerging nationalism, especially after the European shockwave of 1848, which aimed at the political organization of society based on a coincidence of territory with national identity, and in which the territorial separation of people was a means to an end.5 Towards the end of the nineteenth century, nationalist movements became increasingly successful in carving out new, independent nationstates states in the Balkan provinces of the Empire (Greece, Serbia, Montenegro, Romania, Bulgaria, Macedonia, and Albania). With the Empire contracting also in the east, first squeezed by Imperial Russian, French and British expansionist ambitions, and then collapsing in the Ottoman disaster of WW1, the post-war Empire was suddenly left with only Anatolia, the remaining territory, which was proclaimed a nationstate in 1923. During this period, Anatolia had changed profoundly. The Muslim proportion of the Ottoman Empire had stood at around 60% in 1820, a proportion which had increased to 76% by 1890, and 98% by 1923 (Dündar 2001; Zürcher 1998). In a long series of wars between 1783 and 1922 the Ottoman Empire lost the Crimea, Caucasus, and the Balkans, resulting in series of population movements which brought five to six million Muslims to Anatolia.6 Some 1,800,000 Muslims from the Crimea (mainly Tatar) headed to
5 Of the major European states, only the United Kingdom and the Russian and Ottoman Empires managed to escape relatively undisturbed in 1848, when pressure from reformers and radical politicians seeking government changes finally exploded in a series of revolutions which swept across the continent. 6 This was in contrast to the lost Arab provinces, which brought no significant migration to Anatolia, a fact again indicative of the crucial importance of religion (whereas Muslims fled, emigrated or were expelled from Christian Russia and the now Christian Balkan states, the Arab lands were, and remained, Muslim, so there was no compelling reason for the people to move, or be forced to move, when the land slipped out of Ottoman control).
settlement issue and settlement activities
181
Anatolia in the period 1783–1922. Most of them settled in western Anatolia (modern-day İstanbul, İzmir, İzmit, Bandırma, İnegöl and Eski ehir), and mid-Anatolian villages in the Eski ehir-Ankara-Konya triangle. In the period 1859–1879, about 1.5 million Muslim immigrants from the Caucasus were resettled by the authorities in the Black Sea coastal area (Sinop, Trabzon, Samsun, Adapazari, Bolu, and Bilecik). In 1877–78, about 1.5 million Muslim immigrants from the Balkans came to Anatolia and were resettled by the authorities in western Anatolia (Trakya, Bursa, Balıkesir, and Bandırma) and also in eastern Anatolia. In 1912–13 some 640,000 Muslim immigrants from the Balkans were resettled in central and western Anatolia (Tekeli 1990), mainly in the surroundings of Amasya, Tokat, Sivas, Çankırı, Adana, Aydın, İçel, Bursa, Adapazarı and İzmit (Doğanay 1996). In 1923, about 20–25% of the population in Anatolia was of immigrant origin. Various population movements hide behind the aggregated numbers. Alexandros Pallis distinguishes seventeen separate, large population movements between 1912 and 1924 alone, which involved the total transfer of between 2,300,000 and 2,500,000 Greeks, Bulgarians and Turks, in both directions, to and from the two new Balkan states (Box 5.2). Pallis had worked as a Relief Officer in Macedonia in 1913, as Secretary General of the Refugees’ Settlement Commission in Hellenic Macedonia in 1914–15, as Deputy Governor General of Salonika in 1917–18, as Hellenic Commissioner for the resettlement of Eastern Thrace and the Marmara littoral in 1919–20, and as Hellenic Delegate on the Commission for the Exchange of Populations between Greece and Turkey in 19123–24, and had been personally involved with the exchange and settlement of population in this period (Pallis 1925b: 316). His count of migrating peoples thus did not include the loss of hundreds of thousands, maybe as many as two million Armenians, who were removed, one way or another, from Anatolian soil in the East in 1914–1915 under the (Ottoman government) rule of the Committee of Union and Progress.
182
chapter five Box 5.2 Populations Movements to and from Anatolia between
1912 and 1924 (Source: Pallis 1925b: 317–320) 1. 1912. The advance of the armies of the Balkan Allies—the Greek, Serbs, and Bulgarians—resulted in a partial flight of the Muslim population of the invaded area towards Salonika/Thessalonica Of the Muslim population of Greek Macedonia, some 10,000 went over to (still) Ottoman territory as the result of this panic. 2. 1913. On the outbreak of the second Balkan War between Bulgaria and her former allies, a very considerable portion of the Bulgarian population in the districts to the north of Salonika/Thessalonica followed the retreating Bulgarian Army into (what became) Bulgaria. The total number of Bulgarians who migrated at this time was about 15,000. 3. 1913. Towards the end of that year, the whole Greek population of the Macedonian districts ceded to Bulgaria by the Treaty of Bucharest emigrated to Hellenic Macedonia. They numbered about 5,000. 4. 1913. A similar movement took place from the Macedonian districts ceded to Serbia. The Greeks from these districts, to the number of about 5,000 settled for the most part at Salonika/Thessalonica, Florina, and Kilkis. 5. 1913. At the same time the Greek population of the Caucasus, excited by the news of the Greek victories in Macedonia and by reports of free distribution of land, started to emigrate. Although the Greek government, which already had its hands full with other refugees, discouraged the movement, some 5,000 Caucasian Greeks succeeded in being admitted into Macedonia. 6. 1913–1914. As the result of the action of the Bulgarian government in Western Thrace, which territory had been ceded to Bulgaria by the Treaty of Bucharest, and the settlement there of Bulgarian emigrants from Macedonia, practically the whole of the Greek population were forced to emigrate. Of these, some 40,000 settled in Macedonia, with others going to Old Greece. 7. 1914. After the conclusion of peace between the Ottoman Empire and the Balkan states, a considerable portion of the Muslims of central and eastern Macedonia, estimated at 100,000–115,000, left for Ottoman territory, to be settled by the authorities in Eastern Thrace and on the western coast of Anatolia. 8. 1914. About 80,000 Greeks from Thrace and 20,000 from Anatolia were evicted and took refuge in Macedonia, where they were settled by the Greek government. 9. During the First World War (referred to by Pallis in 1925 as the European War) the Bulgarian Army occupied eastern Macedonia, and all the Greek inhabitants—to the number of 36,000—were deported to Bulgaria.
settlement issue and settlement activities
183
10. 1918. Immediately after the Armistice the survivors of the above deportation—to the number of 17,000 only—returned and were reinstated in their homes. 11. 1918–1919. The successive occupations of Western Thrace, Eastern Thrace, and Smyrna/İzmir by the Greek army were followed by the return of the Greeks who had been expelled from these countries in 1913–1914. The total number repatriated from Macedonia was about 140,000. 12. 1919–1920. In the course of 1919 the Greek government decided to remove to Greece the Greeks of south Russia and the Caucasus. Of these, 55,000 were settled in Macedonia. 13. 1919. After the defeat of General Wrangel by the Bolsheviks at Odessa and in the Crimea, a portion of the Russian White Army was transported to Greece. Of these, about 1,000 were settled at Salonika/Thessalonica. 14. 1919–1924. In 1919, a convention was signed between Greece and Bulgaria to facilitate the reciprocal emigration of the Greek and Bulgarian minorities in the two countries. Under this arrangement 27,000 Bulgarians quit Greek Macedonia for Bulgaria up to the end of 1924. 15. 1922–1924. Practically the whole of the Greek population of western Asia Minor and the Black Sea littoral (Pontus) took refuge in Greece. Also, immediately after the signature of the Mudania Convention by which the Allied powers agreed to surrender Eastern Thrace and Constantinople/İstanbul to the Kemalists, the greater part of the Greek and Armenian population of Eastern Thrace and a portion of the Greek inhabitants of Constantinople/İstanbul, fearing reprisals by the Turks, removed to Greece. Of these refugees, up to November 1, 1924, about 200,000 were installed in Macedonia and 120,000 in Western Thrace. 16. 1923–1924. The Greco-Turkish Convention for the exchange of populations was signed in January 1923. Emigration of the Greek and Muslim minorities in the two countries was compulsory, with the exemptions of the Greeks of Constantinople/İstanbul and the Turks of Western Thrace. The convention came into operation in October 1923. By November 1924, the whole of the Muslim population of Macedonia had been transferred to Turkey with the exception of a few individuals of Albanian origin for whom the Greek government had agreed to make a further exemption, amounting to 348,000 people. 17. 1924. In May of this year, the remnants of the Greek population of Eastern Thrace and Asia Minor began to be transferred to Greece, under the provisions of the same convention. By December 1924, when the exchange was practically completed, 150,000 Greeks (94,000 from Anatolia, 18,000 from Eastern Thrace, and 38,000 from Constantinople/İstanbul) had been transferred to Greece. Together with the population that had fled from Anatolia in the years before, their total number was estimated at 1.5 million.
184
chapter five
In (Turkish) social science literature the proposition is maintained that the population movements gave Anatolia its Turkish imprint. (Ari 1995; Karpat 1985; Keyder 1979–1980; Lados 1932; Pallis 1925a; Pallis 1925b). This is, one might say, the received wisdom—and yet it only holds when one equates Turk with Muslim, for if one takes the perspective of language, Anatolia had probably become less ‘Turkish’ by 1923 than it had been in 1913, or 1820 for that matter. Historically, most of the non-Muslims in Anatolia had spoken Turkish—which continued to be the case for the small, non-Muslim minorities (mainly Armenian/ Greek-Christian and Jewish) that remained after the population movements of the first quarter of the twentieth century; many of the Muslims entering Anatolia from the Balkan (mainly Greek) exchanges, however, did not speak proper (Anatolian) Turkish. And if we do not bear in mind the collapse in the meaning difference between religious and a national identity, it is difficult to explain why Turkish speaking Greeks of Anatolia, as well as Christians who were known to have been of Turkish descent, were exchanged with Greek Muslims who hardly knew any Turkish (Akarli 1998: 57). In these cases, considerations of both language and ethnicity seem to have been eschewed for those of religion.7 The ‘Turkish-Greek’ Population Exchange The 1922 Nobel Prize for Peace was awarded to the polar scientist and statesman Fridtjof Nansen (1861–1930), who had interrupted his scientific career in 1905 to urge the independence of Norway from Sweden. After receiving the prize, Nansen was asked by the League of Nations to resolve a complicated dispute between Greece and Turkey, which had displaced most of the Greek-Christian population living in the southern and western coastal areas in Anatolia. Nansen decided that the best solution was a permanent ‘ethnic separation’ and submitted a scheme for the exchange of populations between Greece and Turkey. The resettlement involved the exchange of about 1,250,000 ‘Greeks’ (ethnically Greek-Christians or Christians, mostly ethnically Greek),
7 It seems as if Turkish nationalists such as Ziya Gökalp, Ahmet Ağaoğlu, Yusuf Akçura, and to some extent Ahmet Riza and Abdullah Cevdet, were concerned with turning Muslims into Turks. They considered Muslim religious identity to be a constitutive part of and functional for the development of a Turkish national identity, a notion taken from the positivist Ernest Renan, who himself was a secularist and anti-clerical, but saw religion as a fundamental human need and social bond (Zürcher 2001).
settlement issue and settlement activities
185
who were living on what was now Turkish soil) for about 400,000 ‘Turks’ (Muslims, mostly ethnically Turk), living now in the new state of Greece).8 The immediate background to this decision of Nansen’s, and the early example of intervention and conflict resolution on the part of the forerunner of the UN, was the 1919–1922 Greco-Turkish war, essentially a nationalist battle for control of western Anatolia. Initially, a Greek army had landed in İzmir (then Smyrna) and expanded over a large area of territory in western Anatolia, before retreating again following the pivotal Battle of Sakarya in 1921. The Turks won a decisive victory in August of the following year, 1922, at Dumlupinar, near Afyon, and the Greek army, chased 400km back to Smyrna/İzmir, fled from Anatolia, the local population of Greek (and Armenian) Christians fleeing with it (the Greek and Armenian quarters of Smyrna/İzmir were destroyed in a fire that burned for four days as the Turks took the city). Peace was sealed with the Armistice of Mudyana in October (a few weeks after which the sultanate was abolished and the future course of the Republic of Turkey set). The Greeks had thus already left their Anatolian homeland by the time of Nansen’s proposal for a population exchange, which in essence just accepted their displacement as a fait accompli and sought to regularize the situation by an agreement enabling the proper liquidation of abandoned property. (Ari 1995; Lados 1932; Millas 2001; Pallis 1925a; Pallis 1925b). It might be argued that Nansen’s approach merely recognized the reality on the ground, but the exchange could equally well be roundly criticized for legitimating and extending a separation of populations— and even, in contemporary terminology, for supporting an ‘ethnic cleansing’. The dubious credit for this approach to the problem could not, however, be claimed by Nansen outright. In fact, the idea of a population exchange had first been formulated as early as a decade previously, in a protocol annexed to a 1913 treaty of peace between Bulgaria and the Ottoman Empire. Among other things, this treaty provided for the exchange between the two political units of ‘their’ 8 The term ‘exchange’ assumes the political perspective of the states involved, between which populations move in a two-way direction. For the people concerned, however, the ‘exchange’ obviously tends to be experienced as a one-way movement—a (probably) undesired resettlement over which they typically have little control and in which what they have lost might in no way be compensated by what it is ‘exchanged’ for—and one also that is not necessarily even into the territory, dwellings, etc vacated by their ‘exchangees’ (who they never meet).
186
chapter five
respective populations—Bulgarian (Christian) and Turk (Muslim)— which were resident within fifteen kilometers of the common border, along its entire length (Lados 1932: 18–19). Although the population exchange was nominally voluntary, it was executed on both sides as compulsory. As a result of the exchange, 9,472 Bulgarian families were exchanged with 9,714 Muslim families (Pallis 1925a: 378). A population exchange involving ethnic Greeks/Christians and Turks/Muslims was first considered in 1914, when the Greek PrimeMinister, Eleftherios Venizelos, and the Ottoman Minister of Foreign Affairs, Ghalib Kemal Bey, discussed population exchanges in Thrace, Smyrna/İzmir and the islands in the Aegean See. An exchange was even formally agreed upon by Venizelos and the Grand Vizier, Said Haim Pa a, during a summit in Brussels later that year—but the negotiations were abruptly terminated by the chain of events that led to the outbreak of the First World War in July/August 1914, and the agreement was never settled. It was not until some years later that the idea of a population exchange was finally discussed again and agreed upon, in 1923, after the fighting in Europe and then Anatolia had finished, during the negotiations which culminated in the Lausanne Treaty. The Greco-Turkish convention for the exchange of populations was signed in January 1923 and annexed to the Lausanne Treaty. This agreement was not voluntary, not even nominally so. People were compelled to leave the country upon receiving notice from the Mixed Commission (Karma Komisyonu), a body set up by the Greek and Turkish government specifically to determine persons required to move. From the Greek point of view, it was argued, the solution was on the whole ‘advantageous’. Most of the Greeks had already fled Asia Minor, as explained, and by accepting the compulsory emigration of the remaining population Greece had obtained the right to expel her Muslim population, in whose place the Greek authorities could settle the Greek Orthodox refugees from Turkey. Why, then, did Turkey accept the agreement, which implied that it would be obliged to accept the resettlement of Muslim Greeks, many of them not speaking Turkish, and having no affinity with Anatolia? The answer was demographic: Turkey had lost much of its population in the years of war, and whole districts in Eastern Thrace and Anatolia had become depopulated. Turkey had an interest to repopulate these waste regions as quickly as possible (Pallis 1925a: 380), and instead of a return of the Greek Orthodox, whose political loyalty was regarded as dubious at best, it could settle Muslims in their place. About 80 percent of the ‘Turkish’
settlement issue and settlement activities
187
population expelled from Greece was resettled in areas abandoned by the ‘Greek’ population, i.e. in Eastern Thrace and the Aegean coastal area (Lados 1932: 713). The criterion used to determine whether a person was ‘exchangeable’ or not was religion rather than ethnicity, which is why the categories ‘Turk’ and ‘Greek’ should be placed in inverted commas. The exchange was not between Greeks and Turks as such, but between Orthodox Christians (mostly Greek but also Armenian?) and Muslims (mostly but not exclusively Turkish). The test of religion avoided the thorny questions of whether the Pomaks (Slav speaking Muslims of Macedonia) or the Cretan Muslims, whose mother tongue was Greek, really were Turks. The Mixed Commission had difficulties in determining the persons of Albanian origin, who were exempted from the exchange, an exemption that applied particularly to the Muslim population of Chamouria, a district in Epirus, across from Corfu. By descent these people were Greek Epirotes, but by religion Muslim (having converted to Islam in the 17th century); linguistically they were Albanian, but politically they had been loyal to the Ottoman Empire (having sided with İstanbul against the numerous Albanian insurrections) (Pallis 1925a: 380–81). National Appropriation of Anatolia The population movements resulted in a territorial separation of peoples based partly on ethnic origin, but primarily on religion. In the same period a Turkish nationalist discourse developed, one in which the Ottoman (Anatolian) Muslims were reinvented as ‘Turks’. This term now lost the negative connotations it had had in Ottoman times and took on a positive one.9 As nationalist sentiment developed and finally won the day in Anatolia, the Ottoman/Turkish language also recognized this powerful force. In Ottoman Turkish, the (originally Arabic) word
9 In the high days of the Ottoman Empire, the word Turk was little used, and then in a rather derogatory way. It referred to the Turcoman nomads or, later, the unsophisticated (perceived as ignorant and uncouth) Turkish-speaking peasants of the Anatolian villages. An Ottoman gentleman from Constantinople would have regarded it an insult to be identified as a Turk (in this context, see Chapter 4, Footnote 14). The Ottomans considered themselves (identified themselves as) Muslims. The name Turkey was the name that Europeans had given to Anatolia, a name that only gained currency in the land to which it referred after 1923, when it was adopted as the official name of the new state.
188
chapter five
vatan referred just to the place one was born or one was living until the 18th century, but thereafter came to acquire the meaning of ‘native country’, ‘homeland’ and ‘motherland’, with the connotation of the historical land of ones ancestors. At the end of the 19th century, the (also originally Arabic) word millet, which had been used in Ottoman Turkish to characterize the empire’s religious communities, came to acquire the meaning of ‘nation’ and ‘people’ (from whence milliyetçi and milliyetçilik, ‘nationalist’ and ‘nationalism’).10 And ulus—which had originally been an old Turkish word (ulu ) for territory, ‘country’, and then used by the Mongols (as ulus) to mean ‘empire’ and also ‘nation’ in the sense of a people (which meaning it still carries) and then again ‘reclaimed’ in the Ottoman Empire between the 14th and 17th centuries (with its new, Mongol meaning and corrupted pronunciation)—was resurrected by Atatürk’s Language Society in the 1930s as a ‘Turkish’ alternative to the ‘foreign’ millet, specifically to mean nation, in the modern sense of ‘nation-state’ (and thence, uluslararası, ‘international’, and with the French adjectival -al appended, ulusal, ‘national’) (Mardin 1997; Nalçaoğlu 2002; Lewis 2002). We see in this cursory glance at etymological development of the modern concept of ‘nation’ in the Turkish language, a culmination in the deliberate assignment of a word (ulus) for ‘nation(-state)’, from beginnings two centuries earlier with an expansion of identity (of vatan), from the locality of one’s domicile or birthplace to the wider sense of a shared, historical territory and its common inheritance passed down through time, through the centuries and millennia even. But to what territory did vatan make reference exactly? What was this land of the ancestors? Even though the idea of Turkishness had gained strength during the latter half of the 19th century, at the threshold of the 20th there was still much disagreement and discord about the land the ‘Turkish nation’ had to claim. Most intellectuals referred to as Turkish nationalists today, from Ziya Gökalp to Yusuf Akçura, Ahmet Ağaoğlu, and evket Süreyya Aydemir, were pan-Turkists. For them, vatan referred to Turan, which stretched from southwest China to southeast Europe (the migratory trajectory followed by Turcoman tribes). Anatolia as
10 Even in post-WW1 Kemalist discourse, however, millet and millî (‘national’) seem to have referred to a ‘Muslim nationalism’ (a nation of Ottoman Muslims defined by contradistinction to Ottoman Christians), with little or no direct reference to ethnicity, and separately from politics (the state, devlet), and territory (the fatherland, vatan) (Zürcher 1999).
settlement issue and settlement activities
189
the imagined geographic core of Turkishness was an idea with little pedigree and negligible support. Only few Turkish nationalists took an Anatolian attitude prior to 1923, one of them being Halide Edib. In her article ‘Let us Look After our House’ (‘Evimize Bakalim’), published in the newspaper Vakit of June 30, 1918, she declared Anatolia the first priority, distancing herself from the irredentist dreams of her contemporaries (although she did not openly renounce pan-Turkism, which probably would have been an act of political suicide at the time).11 Realpolitik turned out to be on Halide Edib’s side, however. The ‘national borders’ that were laid down in the ‘national pact’ (misak-i millî), adopted by the last Ottoman parliament in February 1920, were specified by the armistice lines of October 1918, and more or less delineated Anatolia. The territory of the Republic of Turkey was essentially the land that was still defended by the Ottoman army in 1918. After the establishment of the Republic, Mustafa Kemal and his comradesin-arms practically forced the intellectual elite to become Turkish (Anatolian) nationalists. Turkish nationalists adopted Anatolia as their fatherland, but their relationship to Anatolia was hardly passionately patriotic, as reflected in the words of a contemporary journalist: In olden times when we spoke of the nation (millet) we would understand the Turkish [sic] population of Rumelia. The boundaries of the millet would perhaps go as far as the city of Bursa or Eski ehir. Anatolia (i.e. Asia Minor) did not give us a feeling of ‘wholeness’. The regional dialects were so different from one another as to make it difficult for people to understand each other. The people of Trabzon, Konya and Bitlis would not be in tune as was the case for the population of Turks from Salonika, Skopje or Maonastir. Anatolia would be remembered only when people had to be exiled from İstanbul or when another ten thousand men would be sent to their death in Albania or Yemen. (. . .) (F)or Turks Anatolia was the last fatherland (cited in Mardin 1997; 115).12
The loss of Western Thrace and the Balkan provinces, and with it any pretensions to a ‘national’ concept of Rumelia, was a heavy blow to the Ottoman elite. A part of the empire since the 14th and 15th centuries,
11 Author of several articles and novels, Halide Edib was an active member of the Turkish Hearth (Türk Ocak) and Peasantist Association (Köycülük Cemiyeti), both established in 1918. Halide Edib herself established the Peasantist Association, which resembled the narodniki movement in the Russian Empire with its populist approach, focus attention on rural issues and avowed aim of releasing country-folk from the chains of backwardness and ignorance. 12 For the Kemalist pressure on intellectuals, see Zürcher (n.d.).
190
chapter five
the area lay in close proximity to the imperial capital, and also between the empire and the modern West. For two centuries, since the retreat of Ottoman forces into the southern part of the Balkan Peninsular, it had also constituted the border between Islam and Christendom. Indeed, the loss of the empire’s remaining stake in Europe concluded a long decline in its fortunes in the continent and symbolized the passing of former glories, when Hungary had been taken and the walls of Vienna reached in the 16th century during Ottoman zenith under Suleiman the Magnificent (Kanuni Sultan Süleyman). The Janissaries, the crack Ottoman fighting force, had come primarily from Rumelia, as had many, and the most powerful, of the women in the harem. In the nineteenth and into the twentieth centuries, many of the military leaders and top bureaucrats were of Rumeli stock, and even two of the first three presidents of Turkey had their origins there—the first, Mustafa Kemal Atatürk, born and raised in Selaniki/Thessalonica, in today’s Greece, and the third, Mahmut Celal Bayar, whose parents came from Bulgaria. In fact, Rumelia predominated as the place of origin of Turkish nationalist leaders. The typical Turkish nationalist, insofar as there was one, was an Ottoman Muslim man (as a woman, Halide Edib was a rare exception) although of unspecified ethnic background, possibly ethnic Circassian, Kurdish, Arab or Albanian, for example, quite probably from the Balkans, or perhaps Caucasus—but in all likelihood not an Anatolian Turk (Zürcher 2002: 5–6). None of the first presidents were Anatolian Turks—in addition to Mustafa Kemal and Celal Bayar, the second president, İsmet İnönü, was born to a mixed Kurdish-Turkish family from Malatya. Furthermore, when we look at those who, as writers and teachers, contributed to the development of Turkish nationalism, we are struck by the remarkable fact that none of them hailed from an area in Anatolia with an ethnic Turkish majority (Zürcher 2002). As well as Ziya Gökalp, from Diyarbakir in Kurdistan, Yusuf Akçura (1876–1933) was a Tartar from Kazan in the Russian Empire who had participated in the Russian Revolution of 1905, Ahmet Ağaoğlu (1869–1939) was born in the Caucasus, and evket Süreyya Aydemir was the son of migrants from the Balkans.13 Nazim Hikmet Ran, incidentally, regarded by many as the greatest of the republic’s poets and one of the finest exponents of the (modern) Turkish language, was actually the son of a Polish immigrant, Konstantin Borzecki. Borzecki, who had fled Poland after the suppression of the revolution of 1848 in which he had participated, changed 13
settlement issue and settlement activities
191
It is clearly ironic, a paradoxical even, that the movement for an Anatolian nation-state of Turkey was fuelled by individuals who were either non-Anatolian or non-Turk, or both, neither Anatolian nor Turk. This might be accounted for in terms of the observed phenomenon whereby peripheries (of empires) influence the center and thereby determine and define the whole—similarly, for example, to the way in which Prussia on the eastern extent of Germanic Europe came to be at the ideological heart of (and the center of power in) the new Germany (Zürcher 2002b: 1). Geographically peripheral to the old empire, and actually outside the new republic, Rumelia was yet central to the development of Turkey. The phenomenon of Turkish nationalists who were non-Anatolian and/or non-Turk was most obviously embodied in a Rumelian, the towering figure of Mustafa Kemal. After the surname law (Soyadı Kanunu, No. 2525) was passed in June 1934, the man from Salonika/Thessalonica, instigator of the republic and credited with fashioning the nation, was honored (on November 24) by being given the surname ‘Atatürk’, meaning father-Turk—as befitting the founding father, perhaps, but also seeming to imply that there were no Turks before him, which in a sense, of course, was the case (Akarli 1998: 58) Although Atatürk may have been the father of the nation, however, Anatolia was not his fatherland: it was, rather, his ‘stepfatherland’. It is not clear when Atatürk moved from to a genuinely Turkish nationalist stance (as opposed to ‘Muslim nationalist’), but he can certainly be held responsible for moving the capital to central Anatolia (Ankara), having based the resistance movement he led there in 1919. And yet emotionally, however, Mustafa Kemal remained very much attached to Rumelia. There was nothing that pleased and moved him so much as playing and singing old folk songs remembered from the land of his birth, childhood and youth (Zürcher 2002b: 11). The encounters of Turkish nationalism and nationalists with Anatolia were not an unqualified success. Nothing reveals this more than Yakup Kadri Karaosmanoğlu’s novel ‘The Stranger’ (‘Yaban’), a book hailed as example of ‘national literature’ and later to become required reading for Turkish teenagers as part of the standard school curriculum. The main character in Yaban is a doctor who had joined the forces his name in the Ottoman Empire to Mustafa Celaledin and eventually became one of the advisors of the Sultan, rising to the rank of general (pa a). His grandson, Nazim Hikmet later fled himself, escaping Turkey as a communist, after which he lost his Turkish citizenship and became a Polish citizen.
192
chapter five
of Mustafa Kemal in the war of 1918–1923, and who travels with a comrade-in-arms to his village in Anatolia. The visit is big disappointment. Anatolia appears to him as the end of the world: Isn’t there a better place to stay for someone who turned his back on the world than this remote corner in Anatolia? It is as if I am buried alive here (Kadri 1934: 17).14
The doctor’s diagnosis of the village’s inhabitants is infused with repugnancy. [D]isability seems for almost everyone a kind of destiny. The mother of Mehmet Ali is a cripple. One of the sons of Salih Ağa has a hunchback. Zehra, the daughter of Bekir Cavu is blind. I didn’t see it myself, but according to Mehmet Ali, the wife of the muhtar has a disease whose name is not known, that slowly twisted her body into such a terrible contortion that it is not possible to separate her legs from her arms, or her arms from her legs. (. . .) Besides them, there are two crazy people [meczubu] and a dwarf in the village. (p. 19)
He describes the village as an ‘ill and disabled nest’. The characters in the village are without exception repulsive, ready to betray ‘the nation’. The eyes of villagers shine with pleasure after they read the leaflets thrown from airplanes, saying that they will be liberated from the gangs of Mustafa Kemal, and that the Caliphate and Sultan have taken sides with the armies fighting him, while others are ready to give themselves to whoever for a couple of coins: Anatolia . . . The mufti advises the enemy, the village ağa shows the way to the enemy, women hide deserters. . . . How many fighters for the fatherland have been attacked from behind? (p. 110)
The doctor is regarded as a stranger. It makes him full of fury and sadness, as this attitude denies the bonds of nationality: However, later I understood that people from Anatolia, villagers from Anatolia, considered every stranger a ‘Barbarian’, just like the antique Greek named everyone but themselves ‘Barbarians’. ‘One day . . . one day, would I be able to prove them I am not a ‘Barbarian’? The blood in my veins is the blood running in their veins. We speak the same language. All together, we came through the same historical and geographical roads. (. . .) If only I could prove we are the children of the same God! We have the same political fate, the same social bonds; we are related
14
Translations by the author.
settlement issue and settlement activities
193
to each other beyond kinship relations of brotherhood, childhood, and parenthood. (pp. 35–6)
The main character in the book is not able to communicate with the villagers. Their words come from a totally different world, as if the muttering of creatures of a different kind. This is not because they don’t speak Turkish, but because they do not talk like Turks: – – – –
If one is a Turk, how is it possible not to support Kemal Pa a? We are not Turks, sir What are you then? We are Muslims, elhamdülillah. (pp. 152–153)
When the reader filled with disgust and aversion, Kadri turns in anger to the public he had in mind, the Turkish intellectual. The villagers, we now understand, are in fact just victims of their situation, and it is Turkish intellectuals who are held accountable as Kadri asks rhetorically what they have done for their country and their people and upbraids them thus: The Anatolian people had a spirit, which you could not penetrate. It had a mind, which you could not illuminate. It had a body, which you could not nourish. It had a land where it lived! You could not exploit it. You left them in the hands of ignorance, poverty, and famine. (p. 111)
Although Kadri paints a picture of the state of rural Anatolia as a slough of illness, poverty and desperation, he considers it the nation’s only source of power. Anatolia, the author emphasizes, is the true and only fatherland, and the Turkish nation must rise from this arid land, or die. It is the task of the Turkish intellectuals to create this nation: to forge the villages into a national home and to turn peasants into Turks. A monumental task, Kadri acknowledges, since the distance between the urban elite and the rural population was enormous: [T]he difference between a child who has studied in İstanbul and one who has studied in an Anatolian village is greater than that between an Englishman from London and someone from the Punjab in India. As I write these words, my hand is shaking. (p. 36)15
15 Being neither from the Balkans nor the Crimean/Caucasus, Yakup Kadri (1889– 1974) did not himself conform to the norm (or his caricature) of the Turkish nationalist. In fact, he came from an old, aristocratic Ottoman family with vast lands in the area of Manisa, near İzmir—making him a (western) Anatolian Turk by descent—although he was born in Cairo and educated at the French School in Alexandria. Kadri studied law in İstanbul for a brief period in 1908–09, and then became prominent in the New
194
chapter five
Kadri’s Yaban takes its place in a modern tradition of Turkish literature that can be traced back to a short story written during the time of the birth of nationalism (Karabibik/Karabibik by Nabizade Nazim, 1891), which took rural social problems as its primary subject matter, including those of village poverty and ignorance; followed by The Little Pasha (Küçük Pa a by Ebubekir Hazim Tepeyran, 1910), which focused, among other things, on the lack of development in Anatolia and the poor state of village schools and roads (Karaömerlioğlu 2002). The link between the literary theme of rural conditions in the Anatolian hinterland and the development of Turkish nationalism continued and grew during the early republican period, a linkage that gained considerable currency not only in art, culture and the exploration of social philosophy, but also in politics. With the one party (CHP) system operative until 1950, there were few avenues for the expression of alternative views (and even after that, a repressive climate continued to dissuade intellectuals from commenting too directly). Literature has thus had a significant role in shaping Turkey’s political identity, and Yaban had a particularly important effect in the 1930s, the period when rural issues came to the fore with the ‘peasantist’ (köycülük) discourse. As Karaömerlioğlu writes, Yaban ‘sparked great interest among the Turkish ruling elite’, and, furthermore, ‘many historians believe that the interest in village issues that began in the early 1930s is partly due to the impact of this novel’ (Karaömerlioğlu 2002, referring (note 13) to the historical assessments in Carole Rathbun, 1972: 34; and Küçük 1985: 506).16 Literature (Edebiyat-i Cedide) and Dawn of the New Age (Fecr-i Ati) literary movements, which bridged the divide between Ottoman (Divan) and Turkish (National) literature. After going to Ankara to participate in the nationalist independence war, Kadri entered parliament as the representative of Mardin, and later Manisa. Politically a committed Kemalist who personally had a close relationship with Mustafa Kemal, Kadri was the co-founder and publisher of the Cadre (Kadro) journal in 1932, which aimed at the development of a scientific theory of Kemalism (see below), the year also of the publication of Yaban. After Kadro was forced to close its office, in 1934, Kadri embarked on a diplomatic career, as ambassador to several major cities, before eventually returning to parliament to represent Manissa again (1961–1965). 16 Rural realism as a school of literature became even more prominent during the peak of the ‘peasantist’ genre in the 1950s—notably, for example, the 1950 Our Village (Bizim Köy) by Mahmut Makal, a graduate of the Village Institutes, which ‘exerted a shattering impact on political and intellectual circles by dramatically exposing conditions [‘the abject poverty’] in villages’, (Halman 2006: 16). A brutally realistic, sometimes plain grim note has also been struck in the rural life narrative of Ya ar Kemal (e.g. the trilogy, The Wind from the Plain/Ortadirek, 1960/63/68), whose Memed, My Hawk (Ince Memed, 1955) is described by Halman as the ‘apogee’ of the ‘brave new genre’ of the ‘Village Novel’ (ibid.: 17). Ya ar Kemal the renowned writer has also been a political activist—he was imprisoned in 1995 for an article condemning government
settlement issue and settlement activities
195
Building a Nation The Faltering Revolution Yakup Kadri had complained about the faltering of the Kemalist political project, arguing that fashion exhibitions interested the ruling elite more than the crucial problems of the country (Karaömerlioğlu, 1998). Turning ‘peasants’ into ‘the people’ was one such concern. Nusret Kemal Köymen, the founder of rural sociology in Turkey (see Box 5.3), was convinced that the settlement issue was crucial to the fashioning of a Turkish people, for the articulation of a nation. Köymen hailed Settlement Act No. 2510 as a means to societal development (he refers to the law in terms of ‘soy düzeni’, a term we would translate today as ‘racial’ or ‘ethnic order’, but Köymen used the concept of soy in the sense of cemiyet meaning ‘society’).17 In Köymen’s view, Settlement Act No. 2510 would allow for nationbuilding activities based on the principles that 1) cultural and social division and the differential grouping of populations was undesirable, and 2) a new relationship between land and people had to be created (Köymen oppression of the Kurds and supporting the PKK—continuing a cultural tradition of literary engagement in radical politics. 17 Howard Eisenstat (2005) argues that the racial discourse in Turkey in the 1930s was fundamentally designed to act as an inclusionary rather than exclusionary. The concept of race was blended with that of nation and its main thrust directed towards inclusion. The argument is not necessarily wrong, but incomplete. As Eisenstat clearly shows, citing from a inspectorate report from 1940, the existence of a small community of some 500 Armenians, who had done nothing particularly wrong, was nevertheless considered ‘not positive’. The so-called ‘inclusionary’ racial discourse of Turkish nationalism excluded Christians, and was mainly a project of turning Muslims into Turks. Furthermore, one may argue (e.g. following Bourdieu 1992) that all boundaries which act to define, institute and incorporate the whole necessarily accomplish this primarily by exclusion, through an (implicit) definition of what, or who, cannot join and become a member. In stating the accepted we imply the rejected, but more than that, in material terms it is only through rejection of the negative that the positive takes its meaning (that is, even if the intention is inclusionary, the means and thus inevitable effect is exclusionary). Related to the construction of Turkey, the argument would follow that the new nation was actually defined not so much as ‘Muslim’ as ‘non-Christian’. Thus it was that the presence of Jews in the republic, for example, was a matter of relatively little consequence—not because their (religious/cultural/ethnic) identity as Jews was particularly accepted, but rather because it was non-threatening, not Other (so Jews might even be included as ‘nominal Muslims’, insofar as, and precisely because, they were not Christians). (Furthermore, following a Bourdlieun analysis, artificial boundaries are institutionalized through being naturalized: in the case of a nation-state, the population is created as a people, importantly by using the ‘natural’ definer of ethnicity, or, by naturalizing the definition—and thus it was that ‘Turk’ came to replace Moslem (non-Christian) (Bourdlieu 1992).
196
chapter five Box 5.3 Nusret Kemal Köymen
Nusret Kemal Köymen (1903–1964) was a learned thinker and prolific writer. His most passionate and productive years were the 1930s, in particular the period between 1932 and 1936 in which he published the journal Ülkü (Ideal). In his contributions to the journal Köymen critically commented upon reforms and policies that were supposed to contribute to an establishment of the Turkish nation, often expressing unorthodox views. In the June 1936 issue of his journal, Ülkü, published shortly before he left the journal, Köymen wrote: Let us be frank. Beyond doubt our work has been beneficial if compared to doing nothing. However, our most productive work remains uncompleted and carries the risk of not yielding anything. Most of our work is like carrying water to the sea. (Köymen 1936: 299) Born in İstanbul, Nusret Kemal adopted his surname following the 1934 Surname Act. Köymen, meaning ‘village-man’, is a reference to the preeminence that he attached to the village (köy). Others, such as İsmail Hakki Tonguç, the spiritual father of the Village Institutes, preferred to be addressed as ‘Köylü İsmail’. The use of such surnames and forms of address make a political statement, as in common language, the word köy could (and can) have either a positive connotation of nostalgia for home and a simple life, or a derogative one, associated with crudeness and ignorance (indeed, all the negative things picked out by Kadri in Yaban). By identifying oneself with the village, people like Nusret Kemal and İsmail Hakki Tonguç were openly rejecting the negative aspect of köy. Little is known about Nusret Kemal’s early life. He graduated from the prestigious Robert College High School (as, incidentally, did Bülent Ecevit a generation later), which was in Kadıköy, İstanbul, at that time a neighborhood with a relative high number of officers, aristocrats and businessmen, and a population that was religiously and ethnically mixed, including many Christians and Jews. After departing from Ülkü in 1936, Köymen went to the USA to study, graduating from the Department of Rural Sociology at the University of Wisconsin under. Professor John H. Kolb, who had founded the department. It was Kolb who supervised Köymen’s thesis, ‘The Village, the Unit of Societal Organization’, in which Köymen presented what he considered to be the Weltanschauung of the Turkish Revolution. Köymen summarized what he called the ‘tentative synthesis’ of a ‘world outlook, social philosophy, and political program’ in terms of the village as the primary community and basic unit of societal organization. It is important to note here is that Köymen did not consider villages as the seat of agricultural production, but of society as a whole. Cities, with their transient populations, should be the binding centers of intercourse, administration, service and culture for the villages, suggested Köymen, but not the centers of production and population. Therefore, he did not consider
settlement issue and settlement activities
197
Western urbanization to be an inevitable outcome of social evolution, but rather, a developmental mistake (Köymen 1937: 1–2). Köymen was familiar with the work on rural-urban sociology of Pitirim A. Sorokin (1889–1968), the founder of the sociology department at Harvard, and the work on cities of the Chicago school sociologists Robert Park (1864–1944) and Ernest Burgess (1886–1966). He cannot, however, be said to have had very much affinity with their work, which reproduced (instead of bridging) the dichotomy between city and village. Köymen does show a clear affinity with the work of the anarchist Peter Kropotkin (1842–1921). The similarities between Kropotkin’s classic, ‘Fields, Factories, and Workshops’, and Köymen’s later, post-Ülkü work is striking. Both the anarchist Kropotkin and the nationalist Köymen reject the urbanization of industry and were convinced that nations are compelled to find the best method of combining agriculture with industry (Kropotkin 1912; Köymen 1940a). Among others with whom Köymen felt affinity were Henry Maine (1822–1888) and Norman Frederick Grundtvig (1783–1872). The former was an English legal theorist who developed a theory of rural ethnic purity, maintaining that race was found in its purest form in villages. The latter was a Danish church leader who argued that national culture was the greatest power in a nation, and that this was to be found not in cities, but in villages. To Köymen, rural sociology had to study such societal issues as the ‘national being or existence’ and its relationship to the countryside, and not, as rural sociology had developed in the United States, to be concerned just with agricultural production and resources. Most of the work of Nusret Kemal Köymen from the 1930s on, including his thesis at Wisconsin, was centered on a single question, that of how the development of the nation was constituted in the spatial (rural-urban) and economic (agriculturalindustrial) organization of society. In retrospect, Köymen’s later writings are grouped under the general classifications of ‘rural sociology’ (Tütengil 1999) and the ‘sociology of education and learning’ (Üstel 1990), although these disciplines were not very clearly carved out or institutionalized at the time that he was writing. In his work, Köymen turns with ease from issues of education and learning to those of rural development and the integration of agriculture and industry, all for the ‘elevation’ ( yükseltme) and ‘revitalization’ (canlandırma) of the nation. In the 1950s, Köymen’s studies took him into the field of technology and agrarian development (Köymen 1955), and in 1953 he published what can be considered the first Turkish handbook in Educational Sociology (Eğitim Sosyolojsi). He gave lectures at (among others) the National Defense Academy (Milli Savunma Akademisi), the Gazi Education Institute (Gazi Eğitim Enstitüsü), and various teacher training colleges, and was delegated by the government to participate in research and other activities related to international forums, such as Unesco (Üstel 1990).
198
chapter five
1934). As such, Köymen was expressing the ideology of peasantism, which was particularly concerned to find a popular nationalist base for the country in the context of what was, essentially, a state with a rural territory, a conservative culture and an agrarian economy. Recognizing the brutishness of rural poverty and very pressing needs of the people in the era of the Depression, Köymen and the peasantist movement felt that the best direction of development lay in pushing forward and extending the democratization process of the revolution of independent republican nationalism, by further promoting ‘Turkishness’— in the form of equality and homogeneity in the population, through a more profound appreciation and recognition of village life and the ‘real’ Turk in the countryside, and by land reform and new settlement patterns. Köymen thought that Settlement Act No. 2510 provided the means for the development of new settlement type, in particular because it would, he believed, help prevent social differentiation, which he saw as a bigger threat than cultural differentiation, specifically through the development of a new type of agro-industrialized village, where every family would have access to land. New Villages The settlements and organization of the countryside developed into a main concern for Turkish nationalists, and concerted efforts to ‘nationalize’ rural Turkey came to be referred to as ‘internal colonization’ (Barkan 1948; Anonymous 1966). Notions of (cultural) missionaries were invoked, suggesting that the rural population in Anatolia had to be ‘converted’ or ‘civilized’ (Köymen 1939a). And the nationalizing, modernizing conversion of local villagers into national subjects was considered to be contingent on the (re)organization of rural space. The architect Abdullah Ziya emphasized the nationalist mission of architecture in one of his articles in a series on village architecture published in 1933 in Ülkü, arguing explicitly in one that it was the responsibility of architects to design villages that would guide and direct their inhabitants to become Turks (Ziya 1933). For Abdullah Ziya, and many of his contemporaries—Kazim Dirik, Burhan Arif, ükrü Çankaya—the term ‘nation-building’ was more than a figure of speech: they truly believed that the new environments they constructed really could turn villagers into Turks.18 18
The designers of the new model villages tended not, in fact, to be architects
settlement issue and settlement activities
199
The different village plans that emerged from this ‘design for nation’ have in common the application of geometry, which conceptualizes modern space as isotropic (uniform, everywhere the same), corresponding perfectly to the ideal of the nation, which conceptualizes its ‘members’ as if isomorphic units, again, as essentially the same. The homologies in architecture were supposed to contribute to the production of homologies in identity. Burhan Arif ’s model village (1935) consisted of identical village houses laid out on a rectangular, uniform grid that was constructed around a linear shopping street terminating at each end in a square. One square was given administrative functions and contained the buildings of the village director (nahiye müdürü) and gendarmerie; the other was given cultural functions and included such buildings as the school and village museum. Another example of the geometric approach is the ‘ideal republican village’ (1933) of Kazim Dirik, a former general who was inspector general (umûmî mufetti lik) for Thrace. His ideal village was laid in perfect concentric zones for residence, health, education, sports, commerce and industry, all radiating from the center of the circular plan (Bozdoğan 2001: 101–103). Abdulah Ziya (1933) designed an ideal village laid out on a square plan. The site was a perfect square with the village square at the center. Abdulah Ziya did, however, try to combine the universal principles of science (geometry) with vernacular building materials (mud bricks, flat earth roofs), a consideration lacking in the designs of Arif and Dirik. Citing Abdullah Ziya, Sibel Bozdoğan emphasizes that the absence of the mosque, the primary landmark in most Sunni-Muslim villages, was a strong architectural statement, affirming the secularizing agenda of the ruling elite: Edifying and electrifying the villages never occurred to the rulers of the Ottoman Empire, with the exception of some mosques they built in Anatolian villages. These mosques had no other effect than [that of ] strengthening the religious loyalty of the peasants to the sultan and thus turning the oppression of the individual to a collective oppression. The worst thing about a village mosque, which had been the only cultural and social center for the village, is that in its four walls it offers a bastion for the reactionaries who are the organizers of oppression and ignorance. (Bozdoğan 2001: 101)
by profession: Abdullah Ziya had studied at the academy of arts, Kazim Dirik was a general, and ükrü Çankaya was a political sciences graduate and member of the peasantist department of the Ankara People’s House.
200
chapter five
The secular agenda of Turkish nationalists in the 1920s and 1930s has been discussed at length elsewhere (Bozdoğan 2001; Bozdoğan and Kasaba 1997; Zürcher 1998). Our concern here is the relationship between the making of human ecologies, of which the settlement issue is a constitutive element, and the making of the nation. This relationship has two components. The first is the general idea of the nation as a human ecology, i.e. the nation as a habitat, the metaphorical house of Halide Edib’s ‘Let us Look After our House’ (Evimize Bakalım) or Yakup Kadri’s ‘House for Rent’ (Kiralık Konak). The second is the existence and organization of human ecologies within the nation, i.e. the spatial organization of people and settlements within that territorial habitat. These two ideas of the nation and its association with human ecologies are closely related to two different ideas in sociology. The first—the general idea of the nation as a human ecology—involves the idea of the nation as the highest form of human ecologies, and is related to the sociological distinction between ‘culture and civilization’. The second—the spatial organization of people within a territorial habitat—is related to the sociological analysis of industrialization and urbanization. Urbanization and Industrialization History, Köymen insists, shows that the village is the oldest societal organization and probably the oldest form of group life. He also wants us to keep in mind that, historically, both agriculture and industry existed in villages (together or separately), with little of either being situated in cities until recent times (Köymen 1937: 23, 36). Köymen argues that the first cities, the city-states, were by no means self-contained and self-sufficient communities, but rather geographically small, administrative, cultural and economic centers with many consumers and little production. They were the seat of the state rather than the whole of it. The ancient city-states (Köymen only refers to Rome by way of example) declined after they were separated from the component parts, the villages and the bulk of their productive population (ibid. 28–9). Although rural sociology had taken the word village as synonymous with the idea of agricultural community, this was not the inevitable outcome of history, but a peculiar characteristic of development in Western Europe, in which urbanization coincided with industrialization. The village economy may be agriculture-based, but it may perform other
settlement issue and settlement activities
201
economic activities too (forestry, fishery, smuggling and contraband, Köymen enumerates, or gaining income as penitentiary villages, industrial villages, or tourist resorts). The village is not a unit of organization of agricultural production, but of societal organization. For Köymen the decline of the village is strongest in the West (Western Europe and the USA), for which he gives both cultural and material explanations. Culturally, Köymen focuses on religion, viewing the history of the decline of villages as synchronous with the rise of Christianity. Like Marcel Mauss, Köymen regarded society, and in particular national society, as materially and culturally integrated: Köymen proposed that Christianity, with its separation of the spiritual (cultural) from the material, had disrupted this unity. Köymen also mentions the rise of merchant cities and the establishment of manufacturing under the control of guilds as causes for the disintegration of the village as a material and culturally integrated societal cell. It was not that village industry could not compete with the city, quite the reverse, and village cottage-industries, such as weaving in England, Flanders, and India were so resilient that the looms of village-weavers had to be demolished before urban producers could find sufficient markets for their products, and the urban appropriation of production needed to be sanctioned with armed expeditions into (against) villages. Furthermore, as Köymen describes, the effects of the enclosure movement stripped villages of their common land and caused peasants to flock to the cities, and then the Industrial Revolution of the 18th and 19th centuries, sustained by the steam-machine and protected by the state, brought about a further de-industrialization of villages. Heavy and cumbrous machines, placed near urban centers, were unsuited to decentralized production—in contrast, we may note, to the wind and water mills that had driven the earlier, Medieval industrial revolution in Europe, and also, importantly, as Köymen observes, in contrast to the modern generators of electric power (Köymen 1937: 44–48). According to Köymen, the de-industrialization of villages in Anatolia was primarily an effect of the industrial revolution that had occurred in 19th century Europe (Köymen 1939b). Prior to this change, agriculture and industry in Anatolia had been integrated, villages the sites of a variety of industries, and many ‘towns’ nothing but industrial villages. When the Industrial Revolution took off in England, however, handicraft production and exports had collapsed. Water-turbines and steam engines broke the link which formerly had connected farms to industry. Agriculture gradually became the single pillar of the village
202
chapter five
economy. Subsequent governments, moreover, faced with a reduction of exports and export revenues, increased taxes in rural areas. As a result of these two developments, the rural population became all but entirely dependent on agriculture villagers were turned into peasants, forced into working and settling on wasteland and forest land. Households separated from the village and established themselves on the newly acquired land in the proximity, and beyond. Isolated houses and small rural settlements emerged, in areas difficult to access and without decent roads to connect them to each other. The Anatolian settlement structure marked by a small settlement scatter was born out of this process, Köymen thought, which he considered the main cause for social disintegration in the countryside (Köymen 1935).19 Box 5.4 Urban and Rural Modernization in the 1930s:
The case of Kadro and Ülkü The ideas of the rural sociologist and ardent nationalist Nusret Kemal (Köymen) were developed in the midst of political and social upheaval and a search for something new. Warfare and ethnic cleansing in Anatolia had characterized the latter part of the 19th and early 20th centuries; desperate poverty and hunger were widespread. Turkish intellectuals, however, were more concerned with the reality of a nationalist revolution, and frustrated by the lack of an overarching theory to explain its historical specificity and
19 Köymen thus offers a completely different account of the process of hamletization and dispersed settlement structure in Anatolia than that suggested by Ok in his article. However, both accounts do assume the same basic premise of a disintegration from an earlier, more cohesive settlement structure, and, furthermore, given that the cause for fragmentation given by Ok, the Jelali insurrections, predates Köymen’s by centuries, then Köymen’s and Ok’s positions are by no means incompatible. It is not the intention here to consider the historical origins of the settlement structure in Turkey (the brief discussion in Section 4.3.2. suffices for present purposes). Nevertheless, it has to be said that neither Ok’s nor Köymen’s explanation is entirely convincing. Ok’s account may be accurate as a portrayal of an initial dispersive cause, but is surely highly unlikely to be sufficient to account for the situation of Anatolia in 1950, three or four hundred years later—and seems to be conflict with Hütteroth’s (1974) portrayal of inactivity in new settlement construction since before the Jelali insurrections (until a growth towards the end of the 19th century). Köymen’s account may have merits in its depiction of the shift to a rural dependence on agriculture alone, but any increased taxes in rural areas was probably caused less by western Europe’s industrial revolution than by a (re-)centralization of power in the Ottoman empire, and anyway, the Ottoman empire was growing by the latter part of the nineteenth century, led by the agricultural export sector, which does not suggest a picture of rural degeneration and disintegration (Pamuk 2004: 244, 246).
settlement issue and settlement activities
203
prescribe the way forward (Aydemir 1965). In coincidence with a school of thought running from Saint-Simon and Auguste Comte through Durkheim to social-evolutionism and functionalism, these Turkish nationalists were looking for a general theory that could explain the laws underlying the development of the nation and to find a way out of political instability.20 In relation to this quest, two schools of thought were of particular importance, intellectual movements grouped around two different journals: Kadro, published by a group dominated by converted Leninists, and Ülkü, published by Necip Ali and Köymen. The journals represented and reflected opposing radical currents in Turkish nationalist thought that were developed in Turkey of the 1930s. Kadro stuck to the traditional idea of modernization as a process of urbanization in the spatial domain and industrialization in the economic domain, occurring together, in tandem. Ülkü, on the other hand, expressed ideas about a revalidation of modernity, going beyond the urban-rural and agricultural-industrial divide. Kadro’s Urban Industrialism The first issue of Kadro appeared in January 1932 and the final issue, No. 35/36, was published just three years later, in January 1935, the editors closing down the journal after pressure from powerful persons in the CHP (in particular from the prime minister, Recep Peker). During the period of the journal’s publication, its participants had tried to explore a ‘scientific nationalism’, one that would be able to guide the political cadre in their difficult task: the elevation of the nation. Most of Kadro’s subscribers were indeed cadres in the administration, employed in government institutions, especially the Ministry of Education. evket Süreyya (Aydemir) was the architect and theorist of the journal. He had studied social sciences for a brief period in Moscow, a city to which he journeyed in 1921 in the company of Nazim Hikmet. Aydemir returned to Turkey soon after, and in the mid-twenties became a prominent member of the Communist Party of Turkey, publishing theoretical articles on Lenin and Leninism in the journal Light (Aydınlık). Another co-founder of Kadro was Ismail Husrev (Tökin), who had studied economics in Moscow in 1921–1922. Two other key figures in Kadro—Vedat Nedim (Tör) and Burhan Asaf (Belge)—had studied economics and architecture in Germany, influenced there by the Spartakus movement led by Rosa Luxemburg and Karl Liebknecht. In 1926, Tör took over the leadership of the Communist Party of Turkey, but when he ignored directives from Moscow and attempted
20 Comte’s inclination towards the discovery of such universal laws was prompted by political upheaval, the failure of the 1789 revolution to create a new France and the restoration of the monarchy. In Turkey, nationalists too feared a derailing of the revolution, a fear which was fostered by the short lived events associated with the Free Republican Party (August–November 1930).
204
chapter five
to develop a ‘nationalist line’ he was ousted from his position of secretarygeneral and thrown out of the party, together with Aydemir.21 In the course of the 1920s, these founders of Kadro had evolved from Leninists into Turanists. Aydemir frequently lectured at meetings of the Turkish Hearths, a pan-Turkist network which was obliged to close in 1931. Yakup Kadri, who was the intermediary between the Kadro journal and the political center, in particular Mustafa Kemal (Atatürk), was one of the few participants in Kadro who did not have a Leninist background. Yakup Kadri had close contact with Atatürk form the outset, and discussed with him the idea of initiating a journal that would develop and disseminate a theory of national revolution in Turkey (Ertan 1994). The key people in the Kadro movement held that Turkey had experienced a revolution, one that was the product of the first movement of nationalist liberation in the world. This revolution thus held a unique place in world history and could be a guide for other nations on their road to national liberation. Nevertheless, there was no general theory to explain and, more importantly, to determine priorities for policies that would make this revolution more profound (Türke 1999).22 The major proposition of the participants in the Kadro movement was that the struggle between nations was the motor of history and that the industrial revolution had brought this struggle between nations to a new phase, dividing nations as either strong and industrialized, or weak and non-industrialized (Ertan 1994). The idea is a variation of Leninist historical materialism, according to which social development is based primarily on the development of productive forces. The participants in Kadro believed that Turkey had to industrialize in order to be able to catch up with the strong nations. But in order to prevent the emergence of a bourgeoisie and a proletariat, and its product (class struggle, the wedge that splits the nation), the Kadro participants were convinced that industrialization of Turkey had to be accomplished under the total control of the state. Kadro participants did not regard the state as a means to achieve social ends so much as an objective in itself, which, through the process of planning, realizes the nation (Ertan 1994). Kadro was closed due to increasing ideological pressure from the CHP and the government., particularly Recep Peker (Necip Ali, editor of Ülkü, was also influential in his opposition to Kadro.) At that time, Recep Peker was not only the prime minister but also the general secretary of the CHP and, together with Atatürk and Ismet Inönü, a member of the General Presidential Council. He was considered to be most important ideologist of
21 Aydemir and Tör were alleged to have advocated that the party be liquidated and join forces with the Republican Peoples Party CHP. 22 evket Süreyya (Aydemir), Vedat Nedim (Tör), Yakup Kadri (Karaosmanoğlu), Ismail Husrev (Tökin) and Burhan Asaf (Belge) can be considered as the founding fathers of a nationalist or Kemalist left in Turkish politics and political theory, a Turkish predecessor of a tendency in social theory that became known as ‘Third Worldism’.
settlement issue and settlement activities
205
the Party (Ertan 1994; Yıldız 2001: 58–63). Peker was upset by the ideological and theoretical function Kadro had appropriated for itself. Believing that only the party could accomplish the task of formulating the ideology of Kemalism, Peker was displeased by Kadro’s attempts to perform this task outside the party. Necip Ali, meanwhile, blamed the participants in Kadro for distorting the principles of Kemalism by misrepresenting them, in particular by placing the state above the national will rather than considering it the outcome of the national will. Prominent moderates inside the party also, such as Celal Bayar, disliked the radicalism of Kadro. Bayar disassociated himself especially from Kadro’s statism, regarding statism himself as a pragmatic means to and end, the initiation of a developmental process, which includes private enterprise, and not a means in itself (Aydin 2003).23 Ülkü’s peasantism The first issue of Ülkü Halkevleri Mecmuası or simply Ülkü appeared in 1933. The word ülkü, literally ‘ideal’, is generally used to refer to the ideal of the nation—allegedly, Atatürk gave the journal its name, which was also the name of one of his foster children (Aydin 2003: 355; Bakırezer 2004). Ülkü was published by the Ankara People’s House (Ankara Halkevi)—the regime established the People’s Houses in 1932, the year after the closure of the Türk Ocakları youth organization, with the objective of teaching the people the meaning, necessity, and ideals of the Kemalist revolution). The propagation of the principles of the new regime had to create a mass society, which in turn would serve to revive the nation (Karaömerlioglu 1998b). The journal Ülkü had to play a key-role in the diffusion of the values and norms of the Kemalist regime. The inaugural issue of Ülkü stated its mission pathetically: ‘Ülkü is being issued to nurture the spirit of the young generation that has left the dark ages behind and is marching towards a bright future. It is issued to mobilize the revolutionary elements in society . . . and to establish the union of minds, hearts, and action among those committed to this great mission . . . In the writings, analyses, and commentaries of Ülkü, the ideas of republic, nation, and revolution will be primary’ (cited in:
23 Yakup Kadri, looking back, described the demise thus: ‘By the end, even Atatürk himself could not stand the criticisms and rumors about Kadro. What was he able to do as the chief of the state? His party and government did not tolerate us because they did not understand what we advocated. The CHP thought that only it could manage this task. For instance, Recep Peker gave seminars on the Revolution in order to silence our voice’ ( Aydin p116). The appointment of Yakup Kadri as ambassador to Albania was the sign that Atatürk wanted Kadro to cease its activities. Vedat Nedim explains: ‘They [ Peker and others] tried to cut our connection with Çankaya [Atatürk] by having our director Yakup Kadri appointed to Albania’ (quoted in: Aydin 2003: 116). On developmental issues, the participants in the movement remained influential. In 1946, long after the closure of the journal, ex-participants in the Kadro movement contributed to the writing of a Five Year Plan, which sketched a broad outline of a national route for development and set economic priorities.
206
chapter five
Bozdoğan 2001: 93). In contrast to Kadro, which was distributed among the cadre of party and state, the bureaucratic and intellectual elite who resided for the greater part in Ankara, Ülkü was designed as a popular journal, and could be found in the People’s Houses all over the country.24 The journal was an initiative of Recep Peker and Necip Ali, but published by Nusret Kemal Köymen, who must have been appointed to the job for his capacities for theorizing. In 1936, the dominance of this troika over the journal came to a sudden end, when Recep Peker was forced to retire from his post as party secretary-general, and Necip Ali was replaced by Fuat Köprülü as the general director of Ülkü. Nusret Kemal Köymen also had to step aside. The journal was continued, but with a new group of writers (Aydin 2003), and eventually closed down along with the People’s Houses in 1950, when the Republican Peoples Party lost power (Bozdoğan 2001). Politically, the Journal Ülkü was part of the köycülük or ‘peasantism’ current in Turkish politics, and Nusret Kemal Köymen was one of its most ardent adherents. Nusret Kemal Köymen defined peasantism as both a philosophy and a craft (Tütengil 1999), some kind of a philosophy of nationalist praxis. The peasantism current was not specific to Turkey only. The eclectic ideological current had as different representatives as Walter Darre in Germany and Alexander Chayanov in Russia (Karaömerlioglu 1998a). A key element in the ideology of peasantism was the idea that rural life and the peasantry constitute the stronghold of national values and traditions. The most ardent supporters of the peasantist movement even thought that the nation was equivalent to the peasants and made rural society the focus of their interests.25 Kadro vs. Ülkü The participants in Kadro and Ülkü had a common concern: the hampering of the nationalist revolution and the question how to arouse new zeal
24 When they were at their height, at the end of the 1940s, Turkey boasted nearly 500 People’s Houses and about 4,500 People’s Rooms, a smaller village version of the People’s Houses, throughout the length and breadth of the country, although their heart was in the west of the country. 25 Historically, the emergence of peasantist currents went hand in hand with the emergence of Turkish nationalism. In the first publication episode of the influential journal Türk Yurdu (Turkish Land), 1991–1918, the importance of peasant support for the Turkish nationalist ideology and political movement was emphasized time and again. In 1918 (formalized in 1919) the Peasantist Society (Köycülük Cemiyeti) was established, among the founders were many doctors. In narodnik style, its members decided to ‘go to the people’ and settled in villages in Western Anatolia. They not only worked in villages as doctors, but also tried to educate the peasants on Turkish nationalism, and struggled against the means of economic domination by landlords, such as the notorious ‘black books’ listing the debts of peasants and that in fact functioned as a mean of perpetuating serfdom. The peasantists also helped to organize the nationalist resistance movement against foreign invasion. Among them was Dr. Re it Galip, later the Minister of Education in the 1930s.
settlement issue and settlement activities
207
among the intelligentsia, bureaucrats and party officials, and thus mobilize for a program to elevate the nation. Yakup Kadri had complained about the faltering of the Kemalist political project, arguing that fashion exhibitions interested the ruling elite more than the crucial problems of the country (Karaömerlioğlu, 1998). Köymen worried about opinions circulating within the Kemalist elite that it was better not to bother about the fate of the rural population, as that might raise expectations which could not be fulfilled and consequently create hostility against the government (Köymen 1939b). Both wanted to spur on to a new political wave. But the differences between the groups around Kadro and Ülkü were considerable. The persons gathered around Kadro (1932–1935) and Ülkü, in particular in the period the journal was directed by Peker-Ali-Köymen (1933–1936), emphasized and elaborated on different Kemalist ideas, Statism and Populism. A main concern for the participants in Kadro was the question of the relationship between State and Nation, and the need to put a meaning to Kemalist principle of Devletçilik or Statism.26 The journal was discredited by their opponents for the fact that they were allegedly under the influence of communism. However, the Kadro participants were more positivists than Leninists, resembling ideas of Saint-Simon, August Comte, and Ernest Renan. Augusto Comte (1798–1857), the founder of positivism, argued that societies are marked by their own linear progress. The aim of sociology, the science that would hold all other sciences together, was to discover universal laws that govern the organization (social statics) and evolution (social change) of society, a knowledge that could be used to guide social reforms (Seidman 1994). Like Saint-Simon (1798–1857), for whom he had worked as a personal secretary until their split in 1824, Comte envisaged the reorganization of society with an elite of scientists and engineers, leading an orderly process of industrialization that would create a society characterized by harmony. Ernest Renan (1823–1892), who spread the ideas of Comte, canonized the relation between science and progress and believed that the future of mankind was in the hands of an elite of superior minds who were able to understand the ideal and direction of social evolution (Lukes 1973). Intellectually, the participants in Kadro sought to reconcile the theory of Comte with the idea of nationalism, converting Comte’s theory of social change into an evolutionary theory of the development of the nation. In the tradition of Saint-Simon, Comte, and Renan in particular,
26 Statism Devletçilik or was the principle of pre-eminence of the State in the economic field. On the Third and Fourth Congress of the Republican People’s Party in 1931 and 1935, the six ideological pillars of Kemalism were announced: Cumhuriyetçilik or Republicanism, Milliyetçilik or Nationalism, Devletçilik or Statism, Halkçılık or Populism, İnkilapçılık or Reformism, and Laiklik, in general translated as Secularism. Although the significance of each of these concepts in Kemalist thinking is beyond any doubt, the exact meaning and scope is at stake in debates and political struggles between different Kemalist currents (For a discussion of these principles, see Zürcher, 1998).
208
chapter five
the Kadro participants believed that the development of the nation had to be entrusted to an elite with a profound understanding of the laws for national development, not as a transitory stage, but as a condition for development. This might explain why the statism of Kadro appeared as an aim in itself. Like Saint-Simon and the school of that runs from Durkheim to socialevolutionism and functionalism, Comte had relied upon biology for his guiding social imagery. Although the body is an ancient metaphor for political institutions, in the work of the positivists it was used more as a general metaphor for the structure and function of society as a whole. Society was visualized as an organism, and like an organism society was thought to grow from potentiality to self-realization. Body language and the metaphor of the organism were frequently used in Köymen’s publications. In his work, and that of other participants in the Ülkü journal, the idea of people as a body was linked up with that of the nation. This idea of an aggregation of people in a national body had replaced the Ottoman principle that people were organized in separate or autonomous religious communities (millets), only related to each other because they were part of the same territory of the Empire. Now it was thought that, to the farthest corners, people were related to each other not only because they live on the territory of the same country, but because they are part of that same national body. The idea of people as a national body appeared as a crucial conceptual entity. Many of the ideas proposed and discussed in Ülkü were concerned with the aggregation of populations to realize the ideal of the nation. More than anything else, the leading contributors to Ülkü were concerned with defining the principle of Halkçılık or Populism, squeezing the existing multitude into the oneness of the nation.27
The People According to Köymen, the most important feature of social disintegration was the dissolution of ‘the people’ as a social category. Out of a total national population of seventeen million, Köymen reckoned on just ten million as fitting the social category of ‘the people’, calculating that that different forms of differentiation produced seven million people with factional lifestyles, which excluded them from consideration. The seven million figure was made up from two million people regarded as not Turks, five million who were different for socio-economic reasons
27 Halkçılık or Populism was the principle of national solidarity, also occasionally referred to, however, as popular mobilization (not participation), for example in the projects of Peoples Houses and Village Institutes in the 1930s (Zürcher 1998).
settlement issue and settlement activities
209
(one million belonging to the upper-layer of society and two million to the lower-layer), and a further two million people who lived in cities (Köymen 1934d). Socio-economic differentiation was born from a civilizing process in which industrialization and urbanization coincided, Köymen argued. He was particularly concerned that the number of people socio-economically determined as outside ‘the people’ would increase. Köymen’s account underlined the idea that the production of the social category called ‘the people’ is one of the most fundamental problems of nation-states (Balibar 2002: 93). He realized that a state is only able to take the form of a nation to the extent that, through a network of apparatuses and daily practices, the individual is constituted as homo-nationalis. In his 1934 publication, Populism and Peasantism (Halkçılık ve Köycülük), Köymen defined the people as: ‘the majority of people who live within the borders of a country, are bound to each by culture, and have more or less the same level of income and way of life’—thereby combining elements of number, stratification, and culture. Concerning the numerical component, Köymen’s did not expand on the idea that ‘the people’ are those who make up the majority. With regard to the element of stratification, Köymen distinguished three layers or status-groups in society, the upper, middle, and lower layer (specified by both social and material indicators, namely ‘manners’ and ‘income’). As indicated, Köymen considered the ‘upper’ and ‘lower’ layers as extremes differentiated from the ‘the people’, which was itself defined as a ‘middle’ layer only as the result of upward and downward differentiation. Under ideal conditions, ‘the people’ is a non-stratified egalitarian social body (‘in the beginning there was only the people’).28 Köymen argued that in Turkey the two extremes—the lower and upper layer—were not yet much developed in the 1930s. He saw the development of urban-industrialism, as in Western Europe, as inclined to differentiate the populace, with in a small upper layer (bourgeoisie) and a large lower layer (proletariat), as a consequence of which the nation fragments and dissolves. Regarding culture, Köymen distinguished between culture as lifestyle and culture as an expression of ethnicity. Within the aspect of lifestyle,
28 The Turkish word for middle is orta and for common or shared, ortak. Köymen’s analysis seems to collapse the distinction between the middle (orta) layer and a shared (ortak) culture.
210
chapter five
he differentiated between urban and rural, considering rural lifestyles to be compatible with the development of the nation, but urban lifestyles not. Köymen’s conception of the city was bleak. Economically, he compared it with an octopus that strangles villages and exploits their wealth, while socially, he thought that the chaotic life of the cities had given birth to all kind of clubs and associations that had invented their own traditions and provided cohesion for their members, traditions and loyalties that were a threat to the social unity of ‘the people’ as a whole (Köymen 1940c). Köymen did not agree with the general idea in the Western social sciences, that urbanization and industrialization lead not only to a breakdown of customary ways of rural life but also to the production of new modes of collective behavior and social cohesion (which is at the foundation of Durkheim’s distinction between mechanical and organic solidarity, or Smelser’s theory of social integration)—because he considered these new modes of collective behavior factional, not popular, not conducive to social cohesion. Culturally, he argued that the urban lifestyle of cities is degenerative, and thus responsible for its expression in, for example, crime. Köymen was fond of medical metaphors to describe the evil of the city. He compared the city with orthogenesis (Üstel, 1990), defined as the unstoppable growth of an organ, which damages other organs and the body (the nation) of which it is a part (Köymen 1939c). He considered urbanization a pathological process, and warned against what he considered the to be the biggest mistake of Europe, that it had allowed industry to become urbanized. In reference to culture as an expression of ethnicity (soy)—carrying a socio-cultural rather than ethnic connotation—Köymen mentioned the existence of Kurdish, Laz, Circassian, Arab and other village-communities in Turkey (Köymen 1935). From the perspective of building the Turkish nation, he considered the persistence of these villages-communities problematic, being convinced of the need for ethnic identities to dissolve and a new Turkish national identity to be developed (Köymen 1934).29 In reference to experiences in Mexico, where the authorities had established an institute to deal with the integration of the indigenous population, Köymen proposed the establishment of a General Direc-
29 This is still assumed in development projects and policies, see for example the publication ‘trends of social change in the GAP region’ prepared by the Chamber of Agricultural Engineers for the Southeast Anatolia Project.
settlement issue and settlement activities
211
torate for Village Works (Köy İ leri Umum Müdürlüğü) (Köymen 1939b). The Directorate would contribute to the integration of the non-Turkish villagers via a project of alternative civilization, which would ultimately make Turkish citizens of them.30 Culture and Civilization At the basis of Köymen’s critique of modernization is a critique of the separation between culture and civilization, a distinction crucial in the work of Ziya Gökalp (1876–1924). A sociologist and poet, Gökalp was enormously influential during the formative period of Turkish nationalism (above, 1.3.3). In Gökalp’s work the concept of culture is inextricably connected with the concept of the nation. Culture, he argues, marks the boundaries of nations, the highest and most exalted form of political organization. In contrast to the national realm of culture, the concept of civilization is placed in the realm of universal development, marked by science and technology, and for which we would probably use the word modernization today. Both culture and civilization are regarded as open systems, the former open to and accessible through assimilation and the latter by appropriation.31 The distinction between culture and civilization goes back to the work of two thinkers by whom Ziya Gökalp was very much influenced, Emile Durkheim (1858–1917) and his nephew, Marcel Mauss (1872–1950). Gökalp considered Durkheim the most important sociologist of his time. The effect of Durkheim and Mauss on Gökalp can be traced back to an article written in 1913 in which Durkheim and Mauss critically engage with the concepts of nation and civilization. The article starts with the proposition that in studying social phenomena sociologists should
However, Köymen also hints at the empowerment of villagers vis-à-vis the bureaucrats and reformers. He refers to the Mexican experience, where indigenous villagers who have not mastered the Spanish language have a book in which government officials must write down the promises they make. The book stays with the villagers, who can use it to enforce the implementation of these promises. The whole idea seems to suggest an unreliability of the reformer (government official) as a bottleneck in the development of the nation. 31 Born as a Kurd, Gökalp considered himself a Turk. He argued that a person’s national identity is determined not by physical traits but cultural ones. In an illuminating analogy, Gökalp recalls that Alexander the Great said that his real father was not Philip but Aristotle, because the first was the reason for his physical existence while the second for his cultural existence (Gökalp 1922). 30
212
chapter five
take care to relate them to a human group occupying a determinate portion of geographically representable space: ‘It seems then, on first view’, they write, ‘that collective life can develop only within political organisms having definite contours, within strictly marked limits, that is to say, that the national life is the highest form of social phenomenon and sociology cannot know one of a higher order’ (Durkheim & Mauss 1998: 152). Durkheim conceived of the nation as an integrating entity and provider of existential meaning able to inspire a sense of pride and belonging, a human association with the potential to play an integrating role in the modern world: ‘Briefly, the nation was to be more than a collection of groups, sub-groups, and individuals; it also constituted a system of meaning and as such had a sacred character. (Schoffeleers 1978: 5). Durkheim makes a proviso against national reductionism, realizing that the supra-national level (or what may be called ‘international development’) cannot be accounted for in terms of characteristics of separate nations. An analogy is made with collectivities formed by individuals. Interacting individuals, Durkheim posits, constitute a reality that can no longer be accounted for merely in terms of the properties of the individual actors themselves; at this order of reality, the perspective of collective life is necessary, with the nation as the highest form of such collective life. Equally, civilizations are the social facts that encompass several nations, which cannot be understood in terms the properties of the individual nations and must needs be studied at the level of the collective. Durkheim and Mauss refer to Comte, who had made the distinction between a general movement of civilization and its national particularities. Although civilization may assume a peculiar form in any one nation, it was emphasized that the essential elements could not be the product of a state or people alone, but were the product of a higher order. The distinction between nation and civilization allowed Durkheim and Mauss to reconcile the particularistic and parochial idea of the nation with the universal idea of civilization, which necessarily escapes the territorial demarcation that characterizes nations. A civilization does not have well defined limits, but passes beyond political frontiers and extends over a less easily determinable space. A civilization constitutes a territorially open milieu, with each national culture a particular form or expression of the whole (Durkheim & Mauss 1998: 153). Or, in Gökalp’s literary metaphor, ‘the book of civilization consists of national chapters’.
settlement issue and settlement activities
213
Gökalp distinguishes between two types of civilization, city ( ehir medeniyeti) and village (köy medeniyeti) (Gökalp (1923) 1992: 130, 136–9). He argues that in ‘the East’ (what is now east and southeastern Turkey and northern Iraq), Turks were mainly settled in cities, the centers of Turkish culture, and Kurds in villages, the centers of Kurdish culture. Turks who settle in villages Kurdicize, while Kurds who settle in cities Turkify. Following Emile Durkheim and Ferdinand Tönnies, who considered the transition of rural-agricultural communities to urbanindustrial society the most important process of the (our) time, Gökalp argues that modern society is an urban society (and thus, by implication, anticipated the general tendency of a Turkification of Kurds).32 Nusret Kemal Köymen, of course, rallied against this, rejecting the idea that national society is urban. The civilization process as it had evolved—or had been allowed to evolve—in Europe, did not signify a historically determined transition from rural-agricultural communities to industrial-urban society, but instead, produced rural-agricultural communities on the one hand and urban-industrial society on the other hand as two separated social entities (and thereby creating factional interest instead of national unity). Köymen’s basic approach can be summarized in a series of three graphics. Graph 5.1 Representation of the conventional idea of modernization
as a process of transition ( from ‘traditional’ to ‘modern’ society) Urban
Modern Society Agricultural
Industrial
Traditional Society Rural
32 Durkheim encompasses the rural-agricultural/urban-industrial distinction in his mechanical/organic solidarity.
214
chapter five
Graph 5.2 Representation of modernization as a process of differentiation
(between ‘traditional’ and ‘modern’ society) Urban
Modern Society Agricultural
Industrial
Traditional Society Rural
Graph 5.3 Representation of an alternative modernization, as a synthesizing
process in the rural/urban and agricultural/industrial matrix Urban
Modern Society Agricultural
Alternative Modernity
Industrial
Traditional Society Rural
Rurbans Köymen sought a civilization that would reverse the dualism of a disintegration and scattering of rural settlements coupled with the emergence of large cities. He envisaged a new spatial settlement type, which he called ‘rurbans’ (Köymen 1940c). The core of the idea lay in the establishment (or rather, identification) of villages to serve as village-centers (köy-merkezi), centers for other villages providing public and cooperative services. These village-centers would develop nodes in a disconnected scatter of existing settlements and contribute to the development of community feeling. The village-centers were envisaged as collective facilities with communal resources. Basic tools such as hammers and chisels, planes and pliers would be available, along with tap water, Turkish baths (hamam) and showers, plus a small medical clinic, a meeting room, a small shop and a library situated near an administrative
settlement issue and settlement activities
215
office. The idea may be considered an unpolished precursor of the later center-village and village-town models, yet Köymen was convinced that they represented a synthesis of agriculture and industry which would provide the necessary economic foundation for the establishment of the new spatial settlement type, the rurban. Inspired by both the anarchist Kropotkin, to whom Köymen gives credit for inventing the concept of ‘agro-industry’, and the car tycoon Henry Ford, Köymen postulated the idea of the development of village-industries (köycü endüstri) or agro-industries (agro-endüstri) (Köymen 1936). Köymen used the concept of village-industry to refer the establishment in villages of basic industries—simple industrial forms that were essentially upgraded cottage industries, or, forms of craftsmanship with a more ‘business-like’ organization (Köymen 1934d; Köymen 1936b; Köymen 1936c). Agro-industry, on the other hand, encompassed a vertical integration of agriculture and industry, meaning the agricultural production of primary products for industrial processing. Köymen considered Henry Ford a pioneer in the development of village-industries, as he had established twenty village plants near his main factory in Detroit (Köymen 1940b), and also of agro-industries, since he was a key figure in the movement seeking agricultural sources for the primary products that fed industry (Köymen 1940a). These ideas of Nusret Kemal Köymen were not isolated thoughts, it should be added, but took their place in a tradition of social philosophy that challenged the modernization process, arguing that the coincidence of urbanization and industrialization was not only not the only format for progress, but that it was even a wrong one. In this Nusret Kemal Köymen shares his critique of modernization with many different thinkers, including, as well as Peter Kropotkin, Mao Zedong, Haim Halperin, and John Friedmann. Conclusions In this chapter the term resettlement has been used as referring originally to the so-called transfer of populations between different political entities by international agreement, although perhaps also covering informal migration due to or enforced by political changes (such as Muslim Circassians moving to Anatolia following Russian takeover of the territory they had inhabited). That is, ‘resettlement’ was something
216
chapter five
done to people by arrangement between states (be they empires, nations, etc.)—with the proviso that some people may be considered as having ‘actively’ resettled, compelled by political changes although not actually resettled by anyone/anything as such. Tens of thousands of Anatolian Greeks are described falling into this (active) category in reality, having fled Anatolia after the Turkish military victory, although their movement was later formalized by international agreement, thus moving them technically into the first (‘passively’ resettled) category. In referring to this resettlement, the original term muhacir was used in the Ottoman Empire and early Turkish Republic, while Alexandros Pallis used (English language) expressions like exchange of population and racial migrations. During the 1920s and 1930s in Turkey, however, resettlement acquired a very different meaning based on a new objective, that of creating and molding identities. This is illustrated with an interpretation of Settlement Act No. 2510 as pivotal between the old understanding of resettlement as the international transfer of populations between states and the new understanding of resettlement as the fashioning of social life within states. The interpretation of Settlement Act No. 2510 as bridging two understandings of resettlement is discussed through the two backgrounds of the law: legal arrangements to settle Muslim migrants from former Ottoman territories and legal measures for the internal resettlement of Kurds who had rebelled against, or were somehow associated with rebellions against the state. Turkish nationalists welcomed the law as an instrument for the production of ‘the people’. This reflected the new aims of resettlement: not the nationalization of territories by population-transfer and exchange, but the nationalization of people by creating and molding identity. Basically, in the 1930s resettlement was reinvented in Turkey as an instrument for the state to manufacture identities through the reorganization of rural space, a form of nationalist design policy. One current of thought (exemplified by Nusret Kemal Köymen) believed that ‘the people’ was to be understood as a culturally and socially undifferentiated social body, threatened by the fractionizing forces of urbanization and industrialization which created rural and urban sub-cultures. According to this view, only undifferentiated spatial entities (at once both rural and urban, agrarian and industrial) could foster ‘the people’ as a uniform cultural category. This analysis implies a radical critique of the conventional conceptualization of modernization, which equates
settlement issue and settlement activities
217
modernization with the coincidence of urbanization and industrialization (the spatial-economic form modernization took in western Europe and the United States, and which now has since come to part define the concept of the West).
CHAPTER SIX
CLAIMING THE LAND Empirical Observation on the Ground All that is solid melts into air. Karl Marx and Friedrich Engels, Manifesto of the Communist Party, 1848
Introduction During the bitter war with the guerrillas of the Kurdistan Workers’ Party (PKK) in the 1990s, the Turkish army and paramilitary ‘village guards’ systematically evacuated and burned villages in southeastern Turkey. It has been argued that clearance of the countryside and resettlement of the rural population, from which the PKK drew membership, logistical support and intelligence, was a constitutive part of Turkish counterinsurgency. The introduction of the issue of return onto the political agenda in 1995 has been discussed, and how two comprehensive plans aiming at the creation of a new settlement structure were developed in the years 1999–2001, but failed to materialize. Having delved also into the well of nationalist thought and grappled with the historical roots from and with which the civil plans and planners drew and grew their assumptions about resettlement, its objectives and rationale, it is now time to return to the present, and the situation we find ourselves in today. The several return-to-village and rural resettlement projects for the evacuated zones of the Southeast announced by the authorities over the past decade have gone largely unimplemented. According to the analysis presented her, the reasons for this are clear enough: the political situation in Turkey is such that the military have the greater weight of voice in this matter of repopulating the region’s countryside, and militarily it is not welcomed. To paraphrase General Osman Pamukoğlu (Chapter 2, heading quotation), the web that fed the PKK might be broken, but the spider that spun it still lurks. Nationalist Turkey will not be persuaded by the PKK clutching at a ceasefire from the jaws of defeat. However,
220
chapter six
Turkey and its military do not have complete freedom of action here. They are pressured by the UN and UN charters regarding their obligations to the displaced. They are put under the microscope by a variety of non-governmental organizations, both internationally and from within. Political maneuvers, developments and concessions brokered between the EU and successive Turkish governments have combined to add weight to the growing pressure on the army to withdraw politically from the stage. And finally, villagers, families, people, just take matters into their own hands and go back home. In the absence of official return schemes, displaced villagers have been re-migrating by their own means since 1999. Official obstacles to this ‘spontaneous’ return have been documented at great length and in considerable detail by several international human rights organizations, including Human Rights Watch (HRW), the London-based Kurdish Human Rights Project (KHRP), and the Turkey-based Human Rights Association, İHD (İnsan Hakları Derneği), Human Rights Foundation, TİHV (Türkiye İnsan Hakları Vakfı), and Mazlum Der (İnsan Hakları ve Mazlumlar için Dayanı ma Derneği). Among the obstacles to return are a lack of transparent policies and support, denial of permission from authorities, obstruction by the army, fear of and attacks by village guards, and the danger posed by the presence of landmines. Studies by the Turkish Economic and Social Studies Foundation, TESEV (Türkiye Ekonomik ve Sosyal Etüdler Vakfı) not only focus on the issue of return, but also on the need for the authorities to enhance their efforts to address the current conditions of the displaced, mainly since return in any real, full sense is likely to be a lengthy process and remains a distant prospect (Kurban et. al. 2006). Also, the issue of compensation is addressed in relation to the 2004 Compensation Act (Aker et al. 2005, Kurban et al. 2006).1
The Compensation Act—Law No. 5233 on the Compensation of Losses Resulting from Terrorist Acts and the Measures Taken Against Terrorism (Terör ve Terörle Mücadeleden Doğan Zararların Kar ılanması Ar ılanması Hakkında Kanun)—was passed by parliament on July 17, 2004 and became law on July 27, followed by an implementing regulation issued by the Council of Ministers on October 4 and finally amended a year later, on 15 September 2005. The act aims to identify persons who qualify for material damages ‘arising from acts of terror or from measures taken to fight against terror’ (after 1987), and provides for compensation to anyone who has sustained losses due to terrorism or anti-terror activities, including (but not limited to) displaced persons, members of the armed forces, the police and the village guards. It provides for reparation for three kinds of losses: damage to moveable or immoveable property, damage to the life and body of the person, and damage sustained due to the inability to access private property. 1
claiming the land
221
Although an emphasis is placed on displacement, compensation and the right to return, most studies are policy oriented and sociological analyses of return have not been conducted. Of course, the displacement of Kurds has been a subject for sociological study, but research has tended to focus on the living conditions of the displaced after their arrival in the cities, as seen, for example, in the work of Bediz Yılmaz and Ay e Betul Çelik. Yılmaz argues that the displaced Kurds within Turkey are subject to a multidimensional exclusion process. Focusing on a particular neighborhood in inner-city İstanbul (Tarlaba ı), she outlines five forms of exclusion: economic (limited access to formal the labor market); social (limited access to public services, such as health care and education); political (lack of formal political representation); spatial (concentration in particular—poor—neighborhoods), and discursive (in the dominant discourse, the Kurds are depicted as Other) (Yılmaz 2006). In another article, Yılmaz argues that the multilayered mechanism of exclusion hampers upward socio-economic mobility (Yılmaz 2007). Ay e Betül Çelik also links the forced displacement of Kurds to social exclusion, but mainly focusing on the issues of gender and cultural identity, and their representation in civil society organizations, mainly hometown associations.2 Çelik examines how hometown associations
In the affected provinces, damage assessment seven-person commissions composed have been established, six members of which are state-employees. The commissions evaluate requests and assess damage and compensation. But according to Human Rights Watch the Turkish government is failing to provide fair compensation. The human rights organization argues that the provincial damage assessment commissions appear increasingly to apply arbitrary and unjust criteria in calculating compensation, sometimes in defiance of the law’s own implementing regulations. “These calculations appear consistently to favor the government and to be biased against the victims of government abuse” and the law “is being implemented in a manner that directly undermines the stated purpose of the law and the government’s express intent. It also undermines the possibilities for displaced villagers to obtain just compensation and hinders their ability to return to their pre-war homes” (HRW 2006: 3). The Compensation Act suffers from several limitations, not only in respect to its implementation, but also in respect to its content. Article 1, which aims to compensate losses ‘arising from acts of terror or from measures taken to fight against terror’ opens the way to an operative interpretation on the part of the commissions that in practice excludes those persons who were obliged to flee due pressure or effects of the conflict displaced, but not directly evacuated by security forces (or the PKK). The commissions are too linked to the executive rather than the judiciary, and thus subject, apparently and to a higher degree, to state (government and military) pressure. Also, the temporal scope of the act is problematic. Displacement began after the start of the armed struggle of the PKK in 1984, but the coverage of the act dates only from 1987, when the State of Emergency was announced (Çelik, Kurban & Yük eker 2006). 2 Hometown (hem ehri) organisations are associations and foundations regrouping
222
chapter six
express migrant Kurdish women’s problems and identities at the local level and deal with the Kurdish issue at the national level. She argues that increasing contact between Kurds who had come to İstanbul earlier, during the 1970s and 80s, and the newcomers, evacuees and those forced to migrate during the 1990s and after, has had a double effect: a rediscovery of gender identity among the earlier migrants and an increasing consciousness and involvement in the Kurdish issue on the part of hometown associations. She also concludes that forced migration helped existing hometown associations to recognize their Kurdish origin. On the gender issue, Çelik argues that there are important differences. Newly established hometown associations have tried to publicize the issue of forced migration alongside other topics of concern, such as that of gender, but the expression of gender issues had (has) a low priority in already existing associations (established during the seventies and eighties), mainly due to their reproduction of practices such as sex segregation (Çelik 2005). This chapter addresses not the issue of the living conditions in urban areas following forced migration, but the issue of return to the evacuated areas. It is intended to both contribute to an understanding of the meaning of return, and also give evidence regarding the general development of new settlement structures and patterns emerging in the areas evacuated during the 1990s. Although this chapter is mainly concerned with qualitative data, return statistics are also briefly discussed. Qualitative data from four village studies carried out in the province of Diyarbakır disclose a reality behind the bare numbers, and reveal that return takes very different forms. In many cases, it transpires that people do not exchange their urban accommodation for a rural one; instead, the evidence suggests that dual or extended settlement patterns emerge. Furthermore, evidence suggests that not all segments of the population return in equal proportions and that young men and young families in particular are underrepresented among the returnees.
people from the same place, e.g. village, town or district, even province (across Turkey and cutting across educational and economic divides; not specific just to Kurds or the Southeast) (Hersant and Toumarkine 2005).
claiming the land
223
Statistics As described earlier (2.4), the number of settlements evacuated in the war between the PKK and the Turkish state is reckoned at somewhere around 3,000, while the number of people evacuated is put at anywhere between the government figure of about 350,000 (as of 2006, it had been lower), the 2006 Hacettepe University (and semi-official) calculation of 950,000 to 1,200,000, the earlier TESEV estimate of about one and a half million, and that of other NGOs suggesting a number as high as three to four million. Both official and non-official figures indicate that some 93 per cent of the displaced people wish to return to their villages: clearly it makes a vast difference if this percentage is related to 400,000 people, or to four million. The quantitative aspects of return are no less controversial than those pertaining to evacuation, and they are also a little more complex. First, unlike the numbers of the displaced, which are essentially static now, those related to returnees are dynamic, constantly changing as people return. Second, it has been argued by human rights groups in the past that the government figures were inflated (HRW 2005: 17–21), while, ironically, government figures may also contain underestimates (insofar as they are based on official applications to return, for example, those people returning unofficially will not be counted). There are also factors that complicate any simple return/non-return statistical analysis (complex settlement patterns), some of which tend to be particularities, unique, that is, to specific situations and essentially unquantifiable (i.e. statistically insignificant). These, more qualitative factors, are among the phenomena considered in the case studies. Staying with statistics, it is the issue of inflation of the official return figures that is of most importance here. The motivation for any government over-reporting or over-estimation of the numbers of returnees is clear: it would raise the number of returnees as a proportion of the displaced (especially if the number of displaced, the other half of the equation, is under-reported). Turkey’s continuing progress towards full membership of the E.U. is conditional, among other things, on an improved performance in this aspect of human rights.3 It is thus interesting to note that the Turkish government
3
Point 25 of the 1996 parliamentary (draft) report on Turkey’s accession to the
224
chapter six
had given a figure of 124,218 returnees between January 2003 and October 2004 alone (CEC 2004: 51), only 13,418 less than the current figure for the total of all returnees to date. Conclusions derived from this may be various, but a general suspicion as to the reliability of Turkish government statistics must surely be one. In this context, it is instructive to review briefly some of the history of government reporting of return numbers. A close reading of government statistics around the turn of the millennium on return suggests that villagers had been heading home in great numbers. Figures reported by the Regional Development Administration of the Southeast Anatolia Project, GAP (Güneydoğu Anadolu Projesi) indicated that in 1999 alone, 34,509 people (3,593 households) had returned with the support of governors to 97 villages and 82 hamlets, a total of 179 settlements (GAP 2000: 1). However, the emergency region (OHAL) governor, Gökhan Aydıner, had stated on August 7, 2001 that 18,600 people had returned (only a little over half of the number given by GAP two years earlier!). Nevertheless, the Interior Ministry reported three months later, in November 2001, that 30,224 villagers had returned just that year, since 2000. Government sources gave the US different figures again, informing the State Department that 26,481 people had returned by the end of 1999, and 35,513 between June 2000 and December 2001, which made 61,994 returnees. Confidence in this confusion of figures was not helped by a total lack of any accompanying lists of settlements that would have permitted crosschecking—this notwithstanding the precision of the numbers! Civil society organizations claimed that government statistics were not only exaggerated, but also that returnees tended to be village guards (i.e. suggesting that the village return program was not open to everyone). Not only did official figures lack clarity, moreover, data collection by civil society organizations was actively obstructed. For example, in 2002 the Interior Ministry banned a study by the Diyarbakır municipality intended to collect data on the numbers and place of origin of the displaced within the municipality (HRW 2002: 25).
E.U. places the issue of return among its twelve conclusions under the section ‘Human rights and the protection of minorities’. This conclusion (no. 25) welcomes the law on the internally displaced, on the proviso that it be ‘applied efficiently’, and then urges the Turkish authorities ‘to disarm the village guards and to disband the village guard system’, because they are ‘hampering’ return (Eurlings 2006: 9).
claiming the land
225
Focusing on Diyarbakır province, the location of the four village studies (below), statistics collected by the migrant organization Göç-Der indicate that by 2004 more than 45,000 people had returned to 159 settlements (see Table 1). Notwithstanding the claims referred to above, the great majority of these returnees were apparently civilians (and not members of the village guard militias—village guards would get state assistance, but the numbers of villages receiving assistance from the authorities was negligible). It is difficult to determine the proportion of returnees to the number of displaced persons, as we can only guess the exact number of evacuees. Nevertheless, it is possible to make some reasonable judgments regarding proportions of settlements that were evacuated and have since been re-inhabited (re-inhabited, that is, at least in some significant way). Table 6.1 Figures related to village return in the province of Diyarbakir as of
January 1, 2004 (Source: Göç-Der Diyarbakır office inventory) District Centre Bismil Çermik Çınar Dicle Eğil Ergani Hani Hazro Lice Kocaköy Kulp Silvan Total
Re-inhabited villages ? 19 2 10 14 3 10 6 10 32 3 32 18 159
Returning households
Returning population
Villages receiving support from authorities
? 133 40 156 710 18 377 127 121 2,375 137 2,664 224 7,082
? 5,695 175 1,242 8,220 96 2,613 664 665 10,310 872 13,511 1,341 45,404
? 0 0 0 2 (a) 0 1 (b) 0 0 0 1 (c) 1 (d) 0 5
a) Villagers returning to Değirmen (Zixir) and A ağı Haçek received support (details undisclosed). b) Villagers returning to Ergani-Kar ıbağlar (Zilan) received material for the construction of shelters. c) Villagers returning to Kocaköy- aklat ( eqlet) received material for the construction of shelters. d) In Kulp-İslamköy, village guards were settled in a reconstructed hamlet and a newly constructed settlement near the old village. In Hamzalı ( eyh Hamza), 200 of the 258 households that were evacuated returned with support as village guards; in Yaylak (Eskar), 50 of the 300 households returned with support as village guards; and in Zeyrek, 220 of the 280 households returned with support as village guards.
226
chapter six
The human rights organization İHD published a list of settlements that had been evacuated amounting to 432 in Diyarbakır in 1996. I was unable to obtain reliable statistics for the period following this, but since most evacuations occurred in the period between 1991 and 1995, the total number of evacuated settlements in Diyarbakır is not likely to be more than 500 to 600, which, given a re-inhabited figure of 159 villages, would imply the proportion of villages to which people have returned to be somewhere around 25–30%. The Göç-Der statistics further indicate that proportions of the population and households returning to these villages are 35% and 33% respectively of the preevacuation figures. In other words, something approaching a third of all villages were resettled, and within these villages both population and household return was at a similar rate of about a third. Overall, it can reasonably be concluded from these statistics that return in Diyarbakır province is significant, albeit far from high, and almost without any support from the authorities. Methods of data collection The statistics reveal the magnitude of return; however, qualitative methods of research are needed to gain insights in the way return takes place. The main methods of qualitative research that were used to collect data in this study were case studies and field trips. The case studies provided details about return in a particular, clearly demarcated area (a settlement) by use of multiple sources of data (observation, open-ended interviews, secondary sources), while traveling allowed me to collect data in a larger territory. Today, traveling is mainly associated with leisure and tourism, but in earlier times (at the turn of the 20th century) it was an important method for gathering data (among others, in the area under research in this study) and its findings appeared in scientific journals (Maunsell 1894; Yorke 1896; Sykes 1907; Molyneux-Seel 1914; Frödin 1944). Practitioners were in many cases diplomats and military officers, such as the former British military attaché in Constantinople Francis Richard Maunsell, the British honorary embassy attaché and military officer Mark Sykes, or Captain Louis Molyneux-Seel. The work of those men, linked to (British) colonial policy and imperialism, could be described as political geography (Maunsell 1894; Sykes 1907) and anthropology (Seel 1914). I revived traveling as a method suited to the conditions of this study to obtain a bird’s-eye overview on village
claiming the land
227
return (although it was difficult to make estimates of the magnitude of return, it was quite apparent as I passed through on my journeys that village return had occurred throughout the Southeast). For a more detailed study of developments, four settlements in Diyarbakır province were selected. In alphabetical order: Beruk, İslamköy, Matrani, and Mira.4 They are introduced here briefly. On the highway from Batman to Diyarbakır we pass Beruk. The land suffers from a shortage of water, which clearly affects the wheat and lentils that are grown on it. The most important source of income is wheat production and animal husbandry, predominantly sheep. Beruk found itself on a PKK trail and was evacuated and destroyed in 1991. In 2002 the authorities granted permission to three families to settle in the village. These families had never lived in Beruk, but acquired part of the land and found shelter in ruined buildings abandoned in 1991. At the time of my first visit to the village, one year after the families had taken residence, it was hard to find material signs of inhabitation: the power pylons did not carry electricity cables to the village, the water tower was severely damaged, and the houses were still in ruins. İslamköy is located in the Kulp district in the north of Diyarbakır province. Families used to earn a living with arable farming on small plots measured in tenths of a dönüm,5 animal husbandry, and silk production. At the beginning of the 1990s the PKK practically administered the village, until the Turkish army evacuated and destroyed the site in 1994. Today, there is not one İslamköy, but two. One was constructed by the authorities in 2000/2001, on the bare slopes of a mountain not far from the original location of the village. This new İslamköy is inhabited by paramilitary village guards. The other, old İslamköy, is the original village to which the displaced civilian population returns too, in spite of hardship and the opposition of both village guards and authorities. Matrani is located near the Tigris River in Diyarbakır province. The land is divided into plots of medium size, a couple of hundred dönüms per household, and is cultivated with wheat. The wheat is of good quality but water resources are scarce. Animal husbandry is therefore an important additional source of income, as is koruculuk, participation Beruk and Mira are not the real names of the villages. In Turkey, nowadays one dönüm is officially fixed at 1,000 square meters, ten dönüms are one hectare. At different places, however, the area considered to be a dönüm may still vary according to local tradition. 4 5
228
chapter six
in the village-guard militia, and, as later became apparent, the drugs trade. The village is inhabited by Kurds from Viran ehir and Turks from Bulgaria. But most of the Bulgarian settlers have migrated to the west of Turkey, notably the city of Bursa, south of İstanbul. This migration began in the 1970s. The war was an additional reason for the Turkish Bulgarians to join the migration caravan in the 1990s and not a single family returned. Mira is located between the provincial capitals Diyarbakır and Batman. In the distance, mountains rise up impressively, but rolling hills characterize the geography in the immediate surroundings of Mira. Wheat waves in the wind as far as the eye can see. Small streams and brooks cut through the fields. The land—about 25,000 dönüms—was the sole possession of a Kurdish sheikh, but is now divided among his grandsons and great-grandsons. The landowners kept well away from the village as the war intensified and by 1995 had developed an extended settlement pattern. In 2000 and 2001 they returned, but not to the old village. Most of the landowners avoided settlement in the village and settled on their own land, as a result of which Mira developed from a small, but compact settlement into a cluster of hamlets and settlement cores.6 The authorities intended to concentrate the population of the small settlements in a new compact settlement (which they called a village-town) but the idea was abandoned. Each of the cases is an example in itself, telling a story of return. The cases were not selected randomly, but in accordance to similarities and differences (they are at the same time most similar and most different cases). For example, İslamköy and Beruk have in common that they have been evacuated and destroyed in the course of the war between Turkish armed forces and guerrilla fighters of the PKK in the 1990s. The difference between the two villages is that İslamköy is the scene of an official resettlement project, while Beruk was selected as a resettlement site, but implementation failed to materialize as a result of opposition by the displaced villagers. Mira and Matrani have in common that they have not been evacuated and destroyed. Yet most of the inhabitants of Mira left the village as a consequence of
6 The Village Act defines hamlets as settlements with a population of less than 150 inhabitants, but more than 50 inhabitants, or as settlements that are composed of at least five houses and have a minimum population of 30 inhabitants. If an inhabited area does not conform to this standard it is not considered to be a “settlement” but a “settlement core” (iskan çekirdeği), see chapter 4.
claiming the land
229
violence and pressure to become village guards in the second half of the 1990s. Many of them returned after 1999. Matrani experienced little out-migration, but those who left the village did not return. Mira and Beruk are distinguished from Matrani and İslamköy in another characteristic. Mira and Beruk have no paramilitary village guards, but they are present in Matrani and İslamköy. Repetitive ‘mirroring’ between negatives and positives as I traveled brought me to the selection of these four villages (see also table 6.2). Table 6.2 Characteristics of selected villages
Evacuated and destroyed Return (state sponsored) Return (by own means) Village guards Mountainous Plain
Mira
Beruk
Matrani
İslamköy
– – + – + –
+ – + – – +
– – – + – +
+ + + + + –
The cases were selected sequentially. İslamköy was the first on my list. I chose İslamköy because it was evacuated and destroyed in the course of the war, selected as a village return project by the governor of Diyarbakır and a rural development project by the Southeast Anatolia Project GAP (Güneydoğu Anadolu Projesi) in 2001. Beruk was chosen because it had been evacuated and destroyed, and then selected as a re-settlement site, but when this was rejected by its inhabitants, turned into a location to which people returned anyway (without receiving support). I then selected Mira and Matrani, villages that experienced migration but have not been evacuated and destroyed, to contrast with Beruk and İslamköy as evacuated villages. Mental Universe During the course of my research into the villages of Beruk, İslamköy, Matrani, and Mira, I learned that all four had apparently once been settlements inhabited by Christians—mainly Armenians but also Syrians— evacuated either during the anti-Armenian riots of 1895, or in the deportation and killings of 1915, which eventuated in the elimination of the Armenian presence in today’s southeastern Turkey. At the time
230
chapter six
of my first encounters with these settlements, I was not aware that they had been previously inhabited by Christians, but the villagers so frequently referred to tehcir (deportation), katliam (mass killing) or jenosit (genocide) which had resulted in the deportation and annihilation of the Christian population that I was not able to ignore it.7 In my interviews and conversations with local people purportedly about the recent evacuation of the villages in the 1990s by the Turkish armed forces, not only memories were revived, but memories of memories, recollections resonant with old histories of deportation and mass killings. The previous evacuation of their settlements—the evacuation of Christians from the settlements that had occurred in 1895 or 1915)—was part of the mental universe of the Kurdish villagers, who themselves became displaced in the 1990s. Thus I decided to include the issue of the anti-Armenian violence of 1895 and 1915 in Diyarbakir in my village studies, the points of departure being the oral accounts of the villagers I interviewed, which were later cross-checked and verified, or not, with data from secondary sources and archives.8 It is striking that most research on the violent events of the late 19th and early 20th centuries is part of the ultimately unproductive genocide/no-genocide monologues (they tend not to be genuinely comOne would normally expect the word kırm (or soykırm) for ‘genocide’ to be used by Anatolian villagers speaking in Turkish—Turks or Kurds—but it was the word jenosit that I heard. The use of the (imported) term jenosit, evidences a level of politicization among the population, attesting to the success of the PKK in developing a counter power, an alternative political framework of reference to that of the state. Also indicative of this is the use by local people, in some parts at least, of the (imported) term kontra gerilla, rather than the hegemonic (and imported) terörist. Another name for the events of 1915 that I heard used in villages, and in contrast to the (presumably recently) imported jenosit, was an old Ottoman term fermana fila, ferman referring to a decree issued by the Sultan and fila to Christians. The word tehcir is the standard formal term used in Turkey (modern Turkish, derived form Arabic), literally translated as ‘deportation’ or ‘forced migration’, and in practice used almost exclusively to refer to (the evacuation aspect of ) the events of 1915. (The tehcir was formerly decided as Ottoman policy and passed as law in April/May 1915, triggered by the situation in the East which was fast spiraling out of control as anti-Ottoman Armenian forces informally allied to the advancing Russian army took control of Van, and threatened to do so also in Diyarbakir, Erzurum and Bayburt). Issues around the evacuation, the Turkish recognition of mass killings, the relationship of the evacuation to the mass killings and the international definition as ‘genocide’ remain, of course, a matter of some controversy (see below). 8 The revival of the anti-Armenian violence in the Kurdish experiences of the 1990s is a wide theme: the two sets of events are discussed in several publications and articles as if they are part of a single social universe of sorrow and suffering (e.g. see Aslan 2006). 7
claiming the land
231
municative dialogues) and mainly concerned with producing general theories and propositions (Deringil 1998: 69).9 With a few exceptions (e.g. see Wiessner 1997), ‘micro-studies’ have hardly been written. Yet micro-level monographic studies are essential if we are to understand the processes that eventuated in the widespread killing and the destruction of Anatolia’s Armenian population. To fully comprehend what might be termed the ‘local breaking points’, we need to study the (changing) social relations at the local (i.e. village) level (Derengil 1998: 69). Because the study of these village histories was made in the context of my research into village return after the evacuation and destruction of villages in the 1990s, the attempt here to write a microhistory of these villages remains limited in scope. Nevertheless, I was able to collect data that support a reasonably robust argument, strongly indicating that shifting power-relations between local elite groups are to be considered of crucial importance in any understanding of the fate that befell the Christian populations in Diyarbakır province. First, some general features of the historical composition of the population in Diyarbakır province are discussed, followed by an elaboration on the village histories (in the context of Christians, around the turn of the twentieth century), before going on to look at the events of 1895 and 1915 from the wider (local) perspective of Diyarbakır province. Composition of the population in Diyarbakır province: some general backgrounds An examination of the records of Diyarbakır’s population reveals strong fluctuations. Both the total number of inhabitants and the (relative)
9 Recently, for example, the work undertaken by Vahakn Dadrian explicating the genocide argument (e.g. Dadrian 1993; 1999), was questioned by Guenter Lewy (Lewy 2005a), coincidently in the same year as the publication also of Dadrian’s opus on the subject (Dadrian 2005a). The synopsis of Lewy’s argument, presented in The Middle East Qu